Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project
A review of health outcomes for patients admitted out of hours and at the weekend: is there an effect and what are the characteristics and contributing factors? Published by the State of Queensland (Queensland Health), May 2017
ISBN: 978-0-646-97198-8
This document is licensed under a Creative Commons Attribution 3.0 Australia licence. To view a copy of this licence, visit creativecommons.org/licenses/by/3.0/au
© State of Queensland (Queensland Health) 2017
You are free to copy, communicate and adapt the work, as long as you attribute the State of Queensland (Metro North Hospital and Health Service).
For more information contact: Office of the Chief Executive, Level 14, Block 7, Royal Brisbane and Women’s Hospital, Herston Qld 4029 Phone: (07) 3647 9501 Email: [email protected]
An appropriate reference for this publication is: Romeo, M., O’Gorman, R., and Bell, A. (2017). Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project: A review of health outcomes for patients admitted out of hours and at the weekend - Is there an effect and what are the charateristics and contributing factors? Brisbane, Qld: State of Queensland (Metro North Hosptial and Health Services) doi: 10.13140/RG.2.2.33233.94568
An electronic version of this document is available at www.health.qld.gov.au/metronorth/publications
Disclaimer: The content presented in this publication is distributed by the Queensland Government as an information source only. The State of Queensland makes no statements, representations or warranties about the accuracy, completeness or reliability of any information contained in this publication. The State of Queensland disclaims all responsibility and all liability (including without limitation for liability in negligence) for all expenses, losses, damages and costs you might incur as a result of the information being inaccurate or incomplete in any way, and for any reason reliance was placed on such information. Forward
Does a patient admitted out of hours and at the weekend receive the same level of care as someone admitted during the week?
It’s a question healthcare providers have long asked, but there has been no definitive answer. The Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekends (REACH) Project aims to address the challenge of seven-day healthcare delivery from a Queensland perspective through analysis of existing research, safety and quality data, and anecdotal evidence from clinicians.
This systematic review of literature provides a strong foundation for a Queensland wide data-linkage project that further examines the issue with the CSIRO. By understanding the variation in healthcare outcomes, we can examine strategic opportunities to provide the very best care available at any time.
The REACH Project has been an important collaboration between Metro North Hospital and Health Service, Queensland Clinical Senate and Clinical Excellence Division, Queensland Health. It furthers our understanding of the nature of healthcare outcomes for patients admitted to public hospitals out of hours and at weekends.
Through this study, we have found that a number of factors – protective care mechanisms, staffing profiles and system efficiencies – may be key to understanding why one patient has better outcomes than another regardless of their admission day.
I commend the Queensland Clinical Senate and Metro North Hospital and Health Service for undertaking this important piece of work to ensure equitable healthcare access for all Queenslanders.
Ken Whelan Dr John Wakefield PSM
Metro North HHS Deputy Director, Executive Officer Clinical Excellence Division
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project I
Executive Summary
This review was commissioned by the Queensland Clinical Senate in 2015, in partnership with the Clinical Excellence Division, Queensland Health and Metro North Hospital and Health Service, to inform research into health outcomes for patients admitted to Queensland hospitals out of hours and to influence the future development and implementation of best practice models of care. The objectives which guided the review were:
• To identify and analyse the factors that influence inequality of health outcomes to patients admitted/presenting to hospital out of hours and at the weekend.
• To identify the range and scope of services or other interventions specifically targeting inequality of health outcomes of patients admitted/presenting to hospital out of hours and at the weekend.
• To identify care processes that have demonstrably reduced the inequality of health outcomes for patients admitted/ presenting to hospital out of hours and at the weekend.
• To identify the potential application of the review findings in the Queensland setting and areas requiring further research.
Overall, there is mixed evidence of inequitable outcomes for patients admitted out of hours, based predominantly on retrospective analysis of large administrative data sets. Clinical dataset analyses, which were often adjusted for severity of illness, are less supportive that differential outcomes exist. Where evidence does exist, it is limited to a small mortality differences, is condition specific so not generalisable across conditions and fails to address other morbidity or patient reported qualitative outcomes. Since a landmark study in 2001, 4 over 100 papers have explored the difference in outcome across a range of patient populations and health systems.
Published literature in the area identifies a range of human resource, system and process considerations that may improve outcomes for patients admitted at these times. Many contributing factors are identified as being associated with or mitigating against differential mortality for patients admitted out of hours. These were generally hypothesised through the expert analysis by researchers when discussing their findings, rather than definitively tested to determine causality or multifactorial causal linkages. No published pre-and-post intervention studies could be identified, by this review, which specifically examined the impact on mortality that a given intervention, for example timeliness to investigation, treatment, models of care or staffing, may have had if a differential outcome had previously been identified.
There are a number of limitations of the largely descriptive and retrospective studies included in this review making definitive recommendations from the literature difficult to make. In part, the questions that guided the review remain unanswered highlighting a number of gaps in our current knowledge. No studies identified any mechanisms that specifically targeted addressing unequal outcomes as a result of out of hours admission. Any methodological limitations, and the opportunity to explore patient outcomes beyond mortality differences, were noted as part of the review and have provided guidance to the development of a Queensland based data linkage outcome study which is under way. Whilst mortality remains an important quantitative patient outcome differential, any demonstrable differences will be broadened and viewed in the context of patient experience, staffing profiles and non-mortality patient outcome measures.
The authors of this report have been wholly responsible for all data collection, analysis, interpretation and manuscript preparation in consultation with the REACH Project Steering Committee and Research Advisory Group. The findings of the review highlight the gaps in the literature regarding where improvement efforts should be targeted to reduce outcome variation, to then stimulate health services research into the impact of newly implemented interventions that will ultimately better inform strategic planning around the delivery of out of hours care.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project III
Contents
Chapter 1: Background...... 1 Differences in out of hours admissions compared to working hours...... 1
Chapter 2: Criteria for considering studies for this review ...... 2 Types of studies...... 2 Search strategies...... 2 Inclusion and exclusion criteria...... 3 Data extraction...... 3 Quality review...... 5
Chapter 3: The size and consequence of differential outcomes for patients admitted to hospitals out of hours and at the weekend ...... 6 Mortality rates...... 6 Treatment Delay...... 6 Morbidity rates...... 6 Length of Stay...... 6 Readmission rates...... 6 Costs...... 7
Chapter 4: Trends and characteristics of out of hours admissions...... 8 Definitional variation...... 8 Regional (health-system) variation...... 8 Data considerations...... 8
Chapter 5: What is the size and consequence of unequal outcomes for patients admitted to hospitals out of hours and at the weekend? ...... 11 Non-disease specific (unselected) patients...... 11 Medical...... 12 Surgical...... 12 Intensive Care...... 12 Disease specific (selected) patients...... 18 Cardio Respiratory...... 18 Gastroenterology ...... 21 Haematology-Oncology and other medical specialities...... 23 Nephrology and Renal disease...... 24 Neuroscience...... 24 Paediatrics and Obstetrics...... 28 Orthopaedics...... 29 Trauma...... 30
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project V Chapter 6: What influences the inequitable outcome for patients admitted out of hours and at the weekend (when there is one) is undecided...... 32 Patient acuity and/or severity of illness...... 32 Staffing number and seniority...... 33 Treatment delay...... 33 Chapter 7: Which care processes have been effective in reducing unequal health outcomes for patients admitted out of hours and at the weekend?...... 45 Comprehensive Stroke Centres...... 46 Intensive Care Units...... 46 Percutaneous Coronary Intervention and/or Critical Care Units...... 47 Trauma Clinics...... 48 Transplant Teams...... 48 Haematemesis and Melaena Units...... 48 Chapter 8: Reflections, Implications and Conclusions...... 49 Reflections on the literature...... 49 Key findings...... 49 Next steps...... 49
Acknowledgements...... 50 Contribution of authors...... 50 Data sharing statement...... 50 References...... 51 Appendices...... 58
VI Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project Abbreviations
AAA Abdominal Aortic Aneurysm HR Hazard Ratio ACS Acute Coronary Syndrome ICH Intra Cerebral Haemorrhage AE Acute Epiglottis ICU Intensive Care Unit AF Atrial Fibrillation IWH In Working Hours AKI Acute Kidney Injury HDU High Dependency Unit AMI Acute Myocardial Infarction LoS Length of Stay ATC Acute Traumatic Coagulopathy MACE Major Adverse Cardiac Events AusHSI Australian Centre for Health Services MELD Model End Stage Liver Disease Innovation MI Myocardial infarction AVH Acute Variceal Haemorrhage MNHHS Metro North Hospital and Health Services BP Blood Pressure NHS National Health System CABG Coronary Artery Bypass Graft NIHSS National Institutes of Health Stroke Scale CAP Community Acquired Pneumonia NSTEMI Non-ST Segment Elevation Myocardial CCU Critical Care Unit Infarction CHF Congestive Heart Failure OR Odds Ratio CI Confidence Interval OWH Out of Working Hours CMR Crude Morbidity Rate PCI Percutaneous Coronary Intervention COPD Chronic Obstructive Pulmonary Disease PE Pulmonary Embolism CSC Comprehensive Stroke Centre PICU Paediatric Intensive Care Unit CSIRO Commonwealth Scientific and Industrial PR Pulse Rate Research Organisation QCS Queensland Clinical Senate DNT Door to Needle Time REACH Reviewing Equitable Access to healthCare out DRG Diagnosis Related group of Hours and at the weekend ECG Electrocardiogram RR Risk Ratio ED Emergency Department SAH SubArachnoid Haemorrhage EGD Esophagogastroduodenoscopy (upper GI SDH Subdural Haematoma endoscopy) SMR Standardised Mortality Ratio EVH oEsophageal Variceal Haemorrhage STEMI ST-Elevation Myocardial Infarction GI Gastrointestinal TIA Transient Ischemic Attack GWTG Get With The Guidelines (stroke program) T-PA Tissue Plasminogen Activator HDU High Dependency Unit UGIB Upper Gastrointestinal Bleeding HF Heart Failure UGIH Upper Gastrointestinal Haemorrhage Hip # Hip Fracture H&M Haematemesis and Malaena
Figures
Page no. Figure 1 Quality assessment criteria for assessment of abstracts 4 Figure 2 Evidence hierarchy for assessment of the full text articles 4 A Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) flow diagram for Figure 3 5 out of hours
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project VII Tables
Page no. TABLE 1 Summary of main characteristics of included studies 9 TABLE 2 Summary primary papers reviewed and also included within a systematic review paper 10 TABLE 3 Data extraction of studies reporting an impact from admission out of hours with higher mortality or 13 other adverse outcomes – Medical patients TABLE 4 Data extraction of studies reporting no impact from admission out of hours on outcomes – Medical 14 patients TABLE 5 Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Medical 14 patients TABLE 6 Data extraction of studies reporting an impact from admission out of hours with higher mortality or 15 other adverse outcomes – Surgical patients TABLE 7 Data extraction of studies reporting no impact from admission out of hours on outcomes – Surgical 16 patients TABLE 8 Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Surgical 16 patients TABLE 9 Data extraction of studies reporting an impact from admission out of hours with higher mortality or 16 other adverse outcomes – Intensive Care patients. TABLE 10 Data extraction of studies reporting no impact from admission out of hours on outcomes – Intensive 17 Care patients TABLE 11 Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Intensive 18 Care patients TABLE 12 Data extraction of studies reporting an impact from admission out of hours with higher mortality or 19 other adverse outcomes – Cardio Respiratory patients TABLE 13 Data extraction of studies reporting no impact from admission out of hours on outcomes – Cardio 20 Respiratory patients TABLE 14 Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Cardio 21 Respiratory patients TABLE 15 Data extraction of studies reporting an impact from admission out of hours with higher mortality or 22 other adverse outcomes – Gastroenterology patients TABLE 16 Data extraction of studies reporting no impact from admission out of hours on outcomes 23 – Gastroenterology patients TABLE 17 Data extraction of studies reporting mixed outcomes for patients admitted out of hours 23 – Gastroenterology patients TABLE 18 Data extraction of studies reporting no impact from admission out of hours on outcomes 23 – Haematology-Oncology and other medical specialities TABLE 19 Data extraction of studies reporting an impact from admission out of hours with higher mortality or 24 other adverse outcomes – Nephrology and Renal transplant patients TABLE 20 Data extraction of studies reporting an impact from admission out of hours with higher mortality or 25 other adverse outcomes – Neuro Surgery patients TABLE 21 Data extraction of studies reporting an impact from admission out of hours with higher mortality or 26 other adverse outcomes – Neurology patients TABLE 22 Data extraction of studies reporting no impact from admission out of hours on outcomes 26 – Neurology patients TABLE 23 Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Neurology 27 patients TABLE 24 Data extraction of studies reporting an impact from admission out of hours with higher mortality or 28 other adverse outcomes – Paediatrics and Obstetrics patients
VIII Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project Page no. TABLE 25 Data extraction of studies reporting no impact from admission out of hours on outcomes 28 – Paediatrics and Obstetrics patients TABLE 26 Data extraction of studies reporting an impact from admission out of hours with higher mortality or 29 other adverse outcomes – Orthopaedics patients TABLE 27 Data extraction of studies reporting no impact from admission out of hours on outcomes 29 – Orthopaedics patients TABLE 28 Data extraction of studies reporting mixed outcomes for patients admitted out of hours 30 – Orthopaedics patients TABLE 29 Data extraction of studies reporting an impact from admission out of hours with higher mortality or 30 other adverse outcomes – Trauma patients TABLE 30 TABLE 30: Data extraction of studies reporting no impact from admission out of hours on outcomes 31 – Trauma patients TABLE 31 Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Trauma 31 patients TABLE 32 Data extraction of administrative studies identifying patient acuity as a contributory factor of 34 predicting unequal health outcomes for patients admitted out of hours a. Cardio Respiratory patients b. Gastroenterology patients c. Haematology-Oncology and other medical specialities patients d. Medical patients e. Neuroscience patients
TABLE 33 Data extraction of clinical studies identifying patient acuity as a contributory factor of predicting 35 unequal health outcomes for patients admitted out of hours a. Cardio Respiratory patients b. Gastroenterology patients c. Intensive Care patients d. Neuroscience patients e. Surgical (Emergency) patients f. Trauma patients TABLE 34 Data extraction of administrative studies identifying staffing as a contributory factor of predicting 39 unequal health outcomes for patients admitted out of hours’ a. Cardio Respiratory patients. b. Gastroenterology patients. c. Intensive Care patients. d. Medical patients. e. Neuroscience patients. TABLE 35 Data extraction of clinical studies identifying staffing as a contributory factor of predicting unequal 40 outcomes for patients admitted out of hours a. Cardio Respiratory patients b. Gastroenterology patients c. Intensive Care patients d. Neuroscience patients e. Orthopaedics patients f. Trauma patients g. Surgical patients TABLE 36 Data extraction of administrative studies identifying treatment delay as a contributory factor of 42 predicting unequal health outcomes for patients admitted out of hours a. Cardio Respiratory patients b. Gastroenterology patients c. Intensive Care patients d. Neuroscience patients
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project IX Page no. TABLE 37 Data extraction of clinical studies identifying treatment delay as a contributory factor of predicting 43 unequal health outcomes for patients admitted out of hours a. Cardio Respiratory patients b. Gastroenterology patients c. Intensive Care patients d. Neuroscience patients e. Orthopaedics patients TABLE 38 Data extraction of studies analysing specific model – Comprehensive Stroke Centres 46 TABLE 39 Data extraction of studies analysing specific model – Intensive Care Units 46 TABLE 40 Data extraction of studies analysing specific model – Percutaneous Coronary Intervention &/or 47 Critical Care Unit/s TABLE 41 Data extraction of studies analysing specific model – Trauma Clinics 48 TABLE 42 Data extraction of studies analysing specific model – Transplant Teams 48 TABLE 43 Data extraction of studies analysing specific model – Haematemesis & Melaena Units 48
Appendices
Page no. Appendix 1 Project Plan 58 Appendix 2 Search Strategies 65 Appendix 3 Data extraction summary of studies reporting the outcomes for patients admitted out of hours and at 66 the weekend - Mortality Appendix 4 Data extraction summary of studies reporting the outcomes for patients admitted out of hours and at 67 weekend - Treatment Delay Appendix 5 Data extraction summary of studies reporting the outcomes for patients admitted out of hours and at 67 the weekend - Length of Stay Appendix 6 Data extraction summary of studies reporting the outcomes for patients admitted out of hours and at 68 the weekend - Timeframe Appendix 7 Data extraction summary of studies reporting the outcomes for patients admitted of the hours and at 69 the weekend - Region and dataset (administrative & clinical)
Appendix 8 Data extraction summary of studies reporting the outcomes for patients admitted of the hours and at 70 the weekend – DRG and data set (administrative and clinical)
X Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project Chapter 1: Background
Differences in out of hours admissions compared to working hours The challenge of delivering care to achieve equitable outcomes whether patients are admitted during the week, at night or at the weekend has been the focus of a growing body of research. Although higher mortality has been reported for babies born at the weekend than for those born during the week in the USA 1, the UK 2 and Australia 3 since the 1970s, the first investigation of the differential outcomes for patients admitted out of hours in other areas was not reported until 2001. 4
Since then, international research has investigated the possible dangers of out of hours admissions. In more recent times, the issue has become politicised in some jurisdictions and policy makers, health system managers and clinicians have been increasingly interested in the potential risks involved for patients admitted out of hours. The growing body of evidence suggesting outcome differences for patients admitted out of hours remains largely observational, and more recently has matured to include improved risk adjustment methodology, and consideration of patient acuity measures so as to better understand the often contradictory findings.
The aim of the review is to identify whether in fact there is a unequal outcomes for patients admitted out of hours and examines the potential underlying causes of differential health outcomes for patients admitted out of hours. Stemming from this are a range of human resource, system and process considerations that may improve outcomes for patients admitted at these times. More specifically, the reviewers sought to identify any published interventions to address a difference in outcome, had it been present, to obtain a thorough understanding of our current knowledge, practice and experience for health outcomes of patients admitted out of hours.
This synthesis of the literature was commissioned by MNHHS on behalf of the QCS to examine evidence around the risks involved in being admitted to hospital out of hours.
The review is a precursor to, and informs commissioned health outcomes research being undertaken by the CSIRO examining linked hospital data from April 2013 to December 2015 including, workforce, patient experience and death registry data from Queensland. AusHSI has also been engaged to undertake health economic analysis of areas showing differential outcomes to assist in policy formulation for the modification of out of hours health provision across Queensland.
The objectives which guided the review are:
• To identify and analyse the factors that influence inequality of health outcomes to patients admitted/presenting to hospital out of hours and at the weekend.
• To identify the range and scope of services specifically targeting reducing the mortality of patients admitted /presenting to hospital out of hours and at the weekend.
• To identify care processes that have been demonstrated to reduce the inequality of health outcomes for patients admitted/ presenting to hospital out of hours and at the weekend and evaluate their effectiveness.
• To identify the potential application of the review findings in the Queensland setting and areas requiring further research.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 1 Chapter 2: Criteria for considering studies for this review
Types of studies Research studies and systematic reviews conducted from 1995 through to 2015 were considered.
Our aim was to provide a broad overview of the existing evidence base in the context that intervention studies, that reduced the inequality of health outcomes for patients admitted out of hours were the initial focus of our search criteria. The criteria that were applied to structure the review process are discussed below.
We have concentrated on identifying and synthesising the key evidence using a focused, policy relevant framework to keep the tasks relevant and manageable. Framework-based synthesis has been identified as an efficient method for synthesising evidence to inform policy within relatively tight time constraints.5 Due to the lack of intervention studies, per se, any administrative and clinical research for out of hours admissions were then included.
A systemised approach was used to ensure the efficient identification and synthesis of the most relevant evidence. The multiple dimensions covered by the review questions posed a challenge to the process. This challenge was further complicated by the fact that out of hours admission research encompassed a spectrum from population-based literature (for a heterogeneous mix of clinical conditions) to clinically based literature (for specific or mixed patient populations).
As a consequence, there was a huge pool of related literature. The review did not attempt to identify all such related evidence; instead we have used a structured search approach to identify the key evidence. The data extraction and quality assessment, of selected papers, have focused on the most critical information for evidence synthesis rather than aiming to exhaustively extract and critique all the available information in individual papers. We have not appraised the evidence in terms of how future services should be provided nor make recommendations about service configurations.A copy of the project plan that provides detailed description of the review methods is provided in Appendix 1. Search strategies Because of the broad and diverse nature of the review topic, a search strategy was developed and was cycled through a number of iterations in order to maximise comprehensiveness and precision. A variety of search methods were undertaken in order to identify relevant evidence for the review question in a timely fashion. This search aimed to find studies that evaluate the mechanisms that address inequality of healthcare delivery and outcomes out of hours. Key issues for consideration were access to services, appropriate management of patients, service delivery, and models of delivery and clinically appropriate treatment of patients. Limits were applied for time periods specified. Additional search terms were identified from papers discussed with practitioners and the project Research Advisory group. Searches were conducted in two stages with a general search on PUBMED run for each area prior to combining with the below detailed strategies as follows:
Stage 1: general search on PUBMED.
Stage 2: targeted database searches around concepts were:
The condition / domain search strategy of the core areas of out of hours; hospital; study design and outcomes. The general structure of combining following relevant terms was used:
Population: Users of the range of healthcare services presenting or admitted to acute care facilities outside of hours.
Exposure: Delivery of acute care services that addresses the inequality of healthcare outcomes for patients admitted out of hours.
Comparator(s) (control): Users of the range of healthcare services presenting or admitted to acute care facilities within regular working hours.
Outcomes: Processes – appropriateness of level of care, adverse events; Patient outcomes – patient experience and satisfaction, decision making, cost consequences and cost-effectiveness.
2 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project The concepts and possible synonyms (search terms) considered were:
Concept: Out of hours and at the weekend. 24/7 model; 7-day care; 7-day health; 7-day services; after hours; day of week; late in week; night care; night time; night time; off-hours; off shift; out of hours; round the clock; utilisation; weekend; weekend service.
Concept: Hospital & Departments. acute; allied health; care; chemotherapy; elective; emergency department; endoscopy; health care; hospital; intensive care; intracranial haemorrhage; medical; microbiology; myocardial infarction; nursing; operating theatre; pathology; PCI; radiology; rehabilitation; service/s; STI; stroke; surgery/surgical; trauma; unit; unscheduled; unscheduled; ward.
Concept: Study design. cohort study/studies; longitudinal study/studies; follow up studies; prospective study/studies; retrospective study/studies; prevalence study/studies; incidence study/studies; transversal.
Concept: Outcomes. assessment; clinical governance; co morbidity; criteria led discharge; discharge; equity; governance; intervention; medical error; models of care; morbidity; mortality; patient harm; patient safety. Inclusion and exclusion criteria We have included both quantitative and qualitative empirical evidence in the review where relevant. Both Australian and international evidence has been included to ensure that alternative models of out of hours healthcare delivery designed to address the same objectives set out in the QCS Review are considered. We have only included published peer reviewed evidence in order to ensure that we have synthesised evidence that has already undergone methodological and expert scrutiny. Evidence was limited to include the years from 1995 to 2015 to ensure the evidence assessed has context and relevance to current policy and practice, as it was assumed clinical systems and service delivery has substantially changed over time.
We have used a core set of inclusion and exclusion criteria:
Inclusion criteria: empirical data (all study designs); independent variable of interest was out of hours acute health care; report relevant outcomes; written in English and published between 1995 and 2015.
Exclusion criteria: descriptive studies with no assessment of an outcome; opinion pieces and editorials; non-English language papers; conference abstracts.
Studies identified through these searches that were clearly not relevant to the issues under the review were eliminated.Two reviewers independently assessed citations for eligibility. Those thought to fulfil the selection criteria were re-tried in full. Where a judgement could not be made based on the citation, or when consensus could not be reached regarding eligibility the full article was obtained to enable a more comprehensive assessment.
A detailed description of the search strategy is provided in Appendix 2. The search retrieved a large number of results and refinements were made to the search to reduce this number.
Retrieved records were imported into EndNote (Thomson and Reuters). Data extraction A standard data extraction was developed. Data extraction from the papers selected for two reviewers performed the review independently without blinding to authorship or journal publication and results confirmed. If differences were identified they were resolved by discussion.
The approached utilised involved 3 stages:
1. Title (eye-balling): Screened titles for relevance (including check for duplicates) to the aims of the study.
Reference titles were scanned to identify contributions to the definition, criteria, components and conceptual framework of models of care targeting the identified causes of inequitable outcomes for patients admitted out of hours.All searchers were run in PUBMED (incl. MEDLINE); EMBASE and Scopus (via Ovid); Emerald Insight, Cumulative Index to Nursing and Allied Health Literature (CINAHL (incl. PsycInfo); via EBSCOhost) and The Cochrane Library (via Wiley Online Library). Focussed review on the contents of national and international medicine, quality safety, health and economic journals and theses were undertaken with the assistance of ‘advanced’ Google search, Proquest Dissertations and Theses Global. Additional references were identified though examination of references manually from the most recent publications from specific journals and through scrutiny of contents pages of highly relevant journals for the last three years including: British Medical Journal (BMJ); Journal of America Medical Association (JAMA); New England Journal of Medicine, BMJ Quality & Safety; Critical Care Medical Journal; Emergency Medical Journal, Medical Journal of Australia, The Lancet and Canadian Medical Association Journal.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 3 Abstract Review: Included abstracts were then screened for significance (& utility) (impact) to the aims of the study.
Articles included were those that appeared to make a significant contribution to the literature around the health outcomes of those admitted out of hours. That is, for an article to be included it had to provide evidence to either support the background (content and scope), or the factors contributing to inequality of health for people admitted out of hours. It was also included if it provided an analysis of the strategies/ interventions or models of care shown to effect the inequality of health for patients admitted/presenting to the hospital out of hours. Studies that were excluded at this stage were recorded with a short explanation for the exclusion.
Reviewers assessed the abstracts based on the following:
Figure 1: Quality assessment criteria for assessment of abstracts
CATEGORY 1 CATEGORY 2 Publications that contain Publications that detail existing context and background. interventions / models of care, their content and scope.
CATEGORY 6 CATEGORY 3 Other relevant publications Publications that provide that do not align with recommendations for categories 1–5. interventions / models of care that are not evidence-based.
CATEGORY 5 CATEGORY 4 Publications that provide Publications that provide recommendations for recommendations for interventions / models of care that interventions / models of care are evidence-based and derived that are evidence-based. from appropriately structured research methods.
2. Full Text Review: We analysed the full text of publications for their contribution (significance) and included those that met the objectives of the study.
Articles included were those that appeared to have a robust underlying logic from which the investigators were confident that conclusions may be drawn. The assessment of the evidence was performed by two reviewers independently. In case of disagreement consensus was sought via discussion, and in case of further disagreement a third party was consulted. Reviewers assessed the full text format for their contribution towards objectives of the project. Reviewers assessed the full text based on the following:
Figure 2: Evidence hierarchy for assessment of the full text articles
Cohort control studies Systematic Review / Meta-Analysis 1 2 Case control studies RCTs ‘Outcomes’ research Experimental Observational research
Case Studies 3 Other types of 4 Expert opinion without explicit studies (e.g.: interview critical appraisal based,local audit)
This framework has provided a clear structure with which to guide the review while retaining the flexibility that has allowed the development of defining the scope of the search strategies, defining inclusion and exclusion criteria to specify what types of studies will be included.
4 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project All data extraction was carried out directly into summary tables rather than detailed data extraction forms, which would subsequently require summarising. The research was highly heterogeneous therefore a simple, yet inclusive template to summarise the key characteristics and findings from each individual paper was used.The table summarised the study design used, population and setting, main purpose and objectives including outcomes measures and key findings and conclusions (including limitation and implications). Quality review The reviewers were also careful to ensure important contextual information was understood. For example, data definitions, outcome measures, quality of analysis and reporting, the country and health system in which the research was conducted, and whether the research was single-site, multi-site or population based. Existing systematic reviews were also contributory. Literature was then grouped according to themes, both by condition type and for likely factors that may influence out of hours outcomes. A brief narrative commentary was then provided, to complement tabulated summaries, for each grouping, to allow readers the opportunity to explore specific evidence as required.
We have presented the review describing methods, an appraisal and summary of the existing evidence and any evidence gaps identified which are likely to be critical to further development of broader out of hours healthcare delivery strategies.This includes where additional, more detailed, topic specific reviews could be of value or where more primary research is needed, for example on a larger scale to provide definitive evidence on effectiveness.
The results of the review sifting process are given in Figure 3.
Figure 3: A Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) flow diagram for out of hours PRISMA
Records identified through database Records identified via snowballing searches = 12 478 and other searches = 67 IDENTIFIED
Abstracts unable to Records screened by be obtained = 43 title = 1496
Duplicates Removed Did not meet the = 676 inclusion criteria = 40 SCREENING
Category 1 = 40 Abstracts assessed for eligibility = 776 Category 2–5 = 257 Category 6 = 59
Category 2–5 Cardio Respiratory = 26 Full Text Articles Gastroenterology = 15 assessed for eligibility Neuroscience / Neurology = 30 = 257 Haematology = 4
ELEGIBILITY Orthopaedics = 6 Renal = 1 Trauma = 10 Medical = 20 Surgical = 9 ICU = 20
Root Cause of differential outcome = 62 INCLUDED (Before & After Studies) Process mechanisms = 32 Interventions = 0
Reviewing EquitableReviewing Equitable Access Access to healthCare to healthCare outcomes outcomes out of Hoursout ofand Hours at the weekend and at(REACH) the weekend Project (REACH)3 Project 5 Chapter 3: The size and consequence of differential outcomes for patients admitted to hospitals out of hours and at the weekend
The main focus of this review is assessment of the evidence relevant to the QCS Review to identify the underlying causes of differential health outcomes of patients admitted out of hours. However, to provide context we have presented a short overview of the current state of knowledge of the characteristics and drivers that underpin any potential inequality of healthcare outcomes and out of hours. Mortality rates Mortality, which was variably defined (in terms of days post admission) between papers, is the most often reported metric against which a difference is assessed. There is conflicting evidence from the literature examined as to whether mortality is in excess out of hours. There are two principal schools of thought as to whether this reflects lower quality care or other differences in patient acuity and or severity. Care quality is a complex endpoint of human resource, system and process differences that exist Monday to Friday (during the day) to other times. The idea has been supported in the literature by large scale trials, which examine administrative data suggesting that case mix-adjusted mortality rates are higher for patients admitted to hospital out of hours (with most focussing on weekends). 6–18 Other studies seem to debunk the idea identifying mixed or no adverse effects from out of hours admission. 4 19-37 A breakdown of these variables are summarised in Appendix 3. Treatment delay It may be assumed that treatment delays would be associated with adverse outcomes in patients admitted out of hours. Whilst some studies do show an inequitable outcome (morbidity and mortality) 38-54 for treatment delays characterised by delay in admission, delayed evaluation in the emergency department, delays in initial evaluation to patients experiencing longer times to consultation and arrival during shift changes others did not discover the impact as a result of admission timing. 19 20 55-66 A breakdown of these variables are summarised in Appendix 4. Morbidity rates From the limited information available, for patient morbidity, the results are again mixed. A number of studies indicate a discrepancy in patient care, especially for those requiring time-sensitive management, 22 33 59 67-72 care was compromised out of hours to the extent that patients became unnecessarily vulnerable out of hours. Other studies, however, have shown different outcomes depending on the condition for which the patient was being admitted. These discrepancies emphasise the importance in considering the impact of factors such as staffing, 68 73-74 overall hospital activity, 26 48 70 75-79 seniority and experience of staff, 16 21 52 80 access to services 81-84 in addition to patient comorbidities. Length of Stay Length of stay (LoS) is often an indicator for poorer outcomes and increased mortality. Again, the results are mixed with some researchers finding a difference in LoS for patients admitted either out of hours. 10 40 47 50 54 73 85-88 In particular, Freemantle (2012) 10 investigated 30-day mortality and LoS for patients admitted Monday-Friday was 1 day, but for Saturday-Sunday it was 3 days. However, a number of papers 19-21 27 33 35 37 38 56 78 89-94 have not found any differences with the researchers concluding that patients were generally sicker and faced an increased likelihood of stay in-hospital, as well as death, even when severity of illness was taken into account. A breakdown of these variables is summarised in Appendix 5. Readmission rates It is not disputed that the timing of when services are delivered to patients’ care can affect outcomes. Some studies have analysed the consequences of discharges out of hours compared to those during the day, 84 95 however no studies were found as part of the review that investigated the magnitude of readmission rates for patients initially admitted out of hours. It is generally recognised that fewer discharges are made out of hours, perhaps because of decreased staffing or the cross coverage of patients during these times.
6 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project Politics The recent politicisation and system wide policy changes, in some jurisdictions, in response to earlier research that identified a difference in mortality out of hours has further clouded the issue. In particular, public policy attempting to provide a more consistent 7-day health service relied on research finding higher rates of in-hospital death following admission at the weekend compared to Wednesday admissions (1.10 and 1.15 for Saturday and Sunday respectively). 11 The UK government, for example, sought to implement major changes to the industrial framework for clinical staff working in the NHS. Complicating matters further are the questions surrounding what causes differential outcomes, if it does exist; due to the emergence of recent evidence and expert opinion after the policy changes were already put in place. Costs Attempts to quantify the costs and benefits of introducing 7-day services in response to the purported higher death rates for out of hours admissions is beginning to be undertaken in the literature 96 with conclusions that as yet there is little evidence that the planned costs of implementing 7-day services, more broadly, will be cost effective. As such it is likely that further literature will emerge to examine this issue that will ideally guide a more targeted risk based approach to prevent an opportunity cost to other publically funded services.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 7 Chapter 4: Trends and characteristics of outside of hours admissions
The aim of this section is to provide an overview of the current literature around whether patients admitted to hospital out of hours experience differential outcomes compared to similar patients admitted during weekdays.
We identified 146 papers that were included in the review, with the main characteristics summarised in Table 1. The studies fall into two broad groups: those concerned with outcome from out of hours admission for patients admitted with a well-defined diagnostic condition (96) and more general (47) studies. Add to this two relevant systematic reviews; 23 25 and one meta- analysis.45 As such, 117 papers were identified from the primary search and an additional 32 papers met the inclusion criteria secondarily. Formal data extraction was undertaken for all papers meeting the inclusion criteria and included in systematic reviews Table 2 lists these 32 papers and the systematic reviews in which they were included. We have summarised the characteristics of the three systematic reviews Tables 5, 11 and 15 respectively. Definitional variation It is important to note that there is no general consensus as to what actually may be meant by out of hours. Of the papers reviewed, 64 studies 4 7 10-13 15-20 22 23 26 32 37 43 44 46 47 49 50 53 55 58 60 62 64 66 68 71 74-77 81 83 86 90 92 93 107 109-127 compared nights and weekends together; 40 studies 21 28 29 31 33 34 36 42 48 51 52 56 57 59 63 65 67 69 78 82 84 94 98 100 101 103 105 108 128-139 compared nights and weekends separately; and 20 6 9 14 24 27 35 39 40 54 73 80 88 89 91 93 140-144 comparing ‘weekends’ with all other days/times. Five studies 79 99 145-147 compared each day of the week and one study 87 compared ‘day time’ to ‘night time’. The remainder (16 studies), 8 25 38 41 45 72 85 148-156 did not clearly identify an out of hours definition.
These differences in definition of day and time of admission may also partly explain the differences between studies. We have summarised the various timeframe definitions of the 146 studies analysed as part of the review in Appendix 6. Regional (health-system) variation A breakdown of the 146 studies reviewed identified that most prevalent studies were from the USA (62) 9 13 15 16 19 20 22 25-28 31 34 38-40 43 47 50 56 59 60 64 71-73 75 76 78 80 82 83 85-87 90 92-94 99 107 116 121 124 125 127 130 132-134 139 140 142 145 146 148-151 155 followed by the UK (25), 6 7 10-12 14 17 18 35 37 49 53 55 61 63 65 66 68 74 89 115 117 131 147 154 other Europe (21)33 42 51 52 57 67 69 77 79 81 84 98 103 105 110 112 119 122 135 137 141 152 and then Asia (15), 33 41 48 58 62 108 109 113 114 123 143 144 152 156 Canada (12), 4 44 46 54 70 91 100 101 126 129 138 153 and Australia (6), 8 24 36 111 118 120 have also published research of mortality of patient admitted out of hours. This is counter to the perception that the published UK experience is of higher prevalence, given the policy debate and decisions that have been adopted there. A breakdown of these variables is summarised in Appendix 7. Data considerations Of the studies selected from the database searches, the majority were retrospective, observational studies. The review identified that studies finding higher mortality rates for patients admitted to hospital out of hours rely on routine administrative data to adjust for risk of death, but these data may not adequately capture severity of illness. Whilst studies based on routine administrative data use sophisticated methods to account for case mix difference between admissions in traditional working hours compared to those out of working hours, administrative data is heterogeneous and lacks information on patient acuity and complexity of patients on admission. Furthermore, those studies using aggregated data make interpretation for specific disease categories or conditions, as well as geographical variations harder to interpret. Studies using specialist clinical dataset samples where disease or department specific studies collected better patient level data the demonstration of an effect for a mortality difference was either mixed or not apparent. A breakdown of these variables is summarised in Appendix 8. An alternative interpretation for the identification of a differential outcome in administrative studies over clinical studies is the requirement for large sample sizes required to substantiate small outcome differences, which are most often the case when present.
8 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project TABLE 1: Summary of main characteristics of included studies
Characteristics Number of included studies n = (%)
Country Australia 6= (4.1%) Asia 15= (10.3%) Canada 12= (8.3%) Global 3= (2%) Middle-East 2= (1.3%) Other Europe 21= (14.3%) USA 62= (42.5%) UK 25= (17.2%) Study design Retrospective Cohort 130= (89%) Prospective Cohort 12= (8%) Meta Analysis/ Systematic Review 3= (2%) Case Series 1= (0.7%) Setting Cardio Respiratory 26= (17.9%) Gastroenterology 15= (10.3%) Haematology-Oncology and other medical specialties 4= (2.7%) Intensive Care 20= (14.5%) Medical 20= (14.5%) Neuroscience & Neurology 30= (20.7%) Orthopaedics 6= (4.1%) Paediatrics and Obstetrics 5= (3.4%) Nephrology and Renal 1= (0.6%) Surgical 9= (6.2%) Trauma 10= (6.8%) Variables Tested Timeframe (see Appendix 6 for breakdown which analysed time frame admitted) Analysed nights and weekends separately 40= (27.3%) Analysed nights and weekends together 64= (44%) Saturday and Sunday studied 20= (13.7%) Other 22= (15%) Mortality (see Appendix 3 for breakdown of studies which analysed mortality) ‘In-hospital’ (no. of days not otherwise specified) 96= (66%) 2-days 6= (4.1%) 3-days 2= (1.3%) 5-days 1= (0.6%) 7-days 9= (6.2%) 14-days 3= (2%) 15-days 1= (0.6%) 28-days 1= (0.6%) 30-days 24= (16.5%) 60-days 1= (0.6%) 90-days 1= (0.6%) 365 Days 1= (0.6%) Re-admission 2= (1.3%) Treatment Delay (see Appendix 4 for breakdown of studies which also analysed timeframes to treatment) 32= (22%) Length of Stay (see Appendix 5 for breakdown of studies which also analysed patients’ LoS) 25= (17.2%)
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 9 TABLE 2: Summary primary papers reviewed and also included within a systematic review paper
De Cordova, Cavallazzi, Hinds, Primary Study 2012, USA. 25 2010, Global. 23 2014, Global. 97
Ananthakrishnan, 2009, USA. 39 ü
Arabi, 2006, Saudi Arabia. 21 ü
Barba, 2006, Spain. 98 ü
Barnett, 2002, USA. 85 ü ü
Bell, 2001, Canada. 4 ü ü
Button, 2011, UK. 74 ü
Byun, 2012, South Korea. 41 ü
Carr, 2015, USA. 56 ü
Cram, 2004, USA. 9 ü
De Groot, 2012, Netherlands. 42 ü
Dorn, 2010, USA. 43 ü
Ensminger, 2004, USA. 26 ü ü
Gould, 2003, USA. 99 ü
Haas, 2010, USA. 27 ü
Hixson, 2005, USA. 28 ü
Jairath, 2010, UK. 49 ü
James, 2010, USA. 13 ü
Kostis, 2007, USA. 50 ü
Laupland, 2008, Canada. 100 ü ü
Laupland, 2011, Canada. 101 ü
Lee, 2012, Singapore. 102 ü
Luyt, 2007, France. 103 ü ü
Myers, 2006, Canada. 104 ü
Meynaar, 2009, Netherlands. 105 ü
Morales, 2003, USA. 78 ü
Saposnik, 2008, USA. 106 ü
Schilling, 2010, USA. 107 ü
Shaheen, 2009, Canada. 54 ü
Sheu, 2007, Taiwan. 33 ü ü
Uusaro, 2003, Finland. 84 ü ü
Wunsch, 2004, UK. 108 ü ü
Youn, 2012, South Korea. 109 ü
10 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project Chapter 5: What is the size and consequence of unequal outcomes for patients admitted to hospitals out of hours and at the weekend?
In as much as this review has been conducted to provide a sound basis of understanding mortality variation for admitted patients, it is formative in the sense that it will inform subsequent research using linked Queensland public hospital datasets. In identifying the gaps in existing literature and being mindful of the multifactorial nature of the problem MNHHS has engaged the CSIRO to further explore the issue in this context. The types of papers and high level observations supporting or not supporting an effect are outlined below.
Retrospective v. prospective: Of the 146 studies, we have examined in more detail 130 were retrospective, observational studies. 4 6-18 20 22 24 26-29 31-41 43 44 46 47 50-54 56-67 69-76 78-94 98 101 103 105 107 108 110-118 120-126 129-156 Twelve were prospective observational studies. 19 21 42 48 49 55 68 77 84 100 119 128 The remaining four studies being two systematic reviews, 23 25 1 meta-analysis 45 and 1 case series. 99
Specific v. non-specific disease types: Ninety-seven papers 19 20 38-43 46 55-57 68 73-76 80 81 89 110 128 129 131-134 148-152 13 18 27 34-36 44 45 47-54 58-66 69-72 77 82 83 86 88 90-94 99 109 112-116 118 119 121-123 125 126 135 136 138 140 142-146 153 154 156 analysed specific patient cohorts (cardio respiratory, gastroenterology, haematology, nephrology, neurology/neuroscience, paediatrics and obstetrics, orthopaedics and trauma). The remaining 49 papers 4 6-12 21-26 67 85 98 111 130 14-17 28 29 31-33 37 78 79 84 87 93 100 101 103 105 107 108 117 120 124 127 137 139 141 147 155 analysed non-disease specific aggregated data.
Multi-site v. Single-site: One hundred and five (studies were multi-site studies 4 6-20 22 24 31 32 35 36 38 39 43 44 46-51 53-60 62 64 66 68-76 79-82 84-86 88-94 99-101 103 105 107-111 115-117 119 121 122 124-126 128 130 131 135-140 142-146 151-153 155 157 with 70 of these analysing administrative datasets. 4 6-20 22 24 32 38 39 43 44 46 49-51 53 54 57 58 60 64 66 69 71 73-76 80 81 85 86 89 90 92 93 107 111 113 116 117 122 124-127 135 138-144 146 147 149-153 155
Relationship between a ‘positive’ study and administrative nature: We identified that of 60 studies 6-18 42-46 48 50 52-54 68 69 72-74 76 77 79 80 82 85 86 88 93 98 107 111 116 118 119 122 125 134 137 139-141 143 146 149 151 152 154 155 reporting higher mortality outcome, 45 6-18 43 44 46 50 53 54 69 73 74 76 80 85 86 88 93 107 111 116 122 125 126 139-141 143 146 149 151-153 155 of these studies analysed an administrative data set.
Relationship between a ‘positive’ study and clinical nature: We identified of the 66 clinical studies 21 26-29 31 33-37 40-42 47 48 52 55 56 59 61-63 65 67 68 70 72 77-79 82-84 87 91 94 98 100 101 103 105 108-110 112 114 115 119-121 123 128-134 136 137 145 148 154 156 reviewed, 40 papers 21 27-29 31 35-37 40-41 47 56 59 61-63 65 67 70 78 83 87 91 94 101 103 105 108-110 114 120 128-129 132-133 136 145 148 156 found little or no significant difference in mortality outcome by day of admission. A breakdown of the non-disease specific and specific specialised cohort patient group studies is discussed in more detail below.
These studies have shown that an effect does not apply to all diagnoses. Mortality associated with out of hours has been reported by a number of authors across a spectrum of diseases, being most evident when diseases are studied within a system at the same time. However, for every paper supporting the presence of increased mortality rates with out of hours admission, another paper can be found that disputes this finding. This is most evident in the research surrounding stroke and AMI where there are a number of papers supporting and refuting the presence of an effect for patients admitted out of hours. Further work is needed to clarify the cause of these outcomes and in particular to distinguish the patient and pre-hospital factors from the in- hospital factors. Non-disease specific (unselected) patients We identified 49 relevant studies 4 6-12 14-17 21-26 28 29 31-33 37 67 78 84-85 87 93 98 100 101 103 105 107 108 111 117 120 124 127 130 137 139 141 147 155 from the database searches that have been categorised as analysing the outcomes for non-specific patient cohorts. Of these 49 non-disease specific (unselected) patient group studies, 20 papers 4 6 8-12 16 22 24-25 37 98 107 117 124 127 139 141 147 addressed medical patient cohorts; nine papers 7 14-15 17 31 32 87 93 155 analysed surgical patient cohorts and 20 papers 21 23 26 28 29 33 67 78 84 85 100 101 103 105 108 111 120 130 137 analysed intensive care patient cohorts.
Seventeen papers 4 9 21 26 28-29 33 78 84 85 98 100 101 103 105 107 108 of which were included in two systematic reviews. 23 25 These systematic reviews are not included in the summary tables.
The papers examined have variable results. Thirty-three of these studies 6-12 14-16 22 24 31 32 79 84 85 100 101 103 105 108 111 117 124 127 130 137 141 147 included multiple sites with 23 6-12 14-16 22 24 32 85 93 111 117 124 127 139 141 147 155 of these using administrative datasets in their analyses. Of the 21 studies examining clinical datasets, 21 26 28 29 31 33 37 67 78 79 84 87 98 100 101 103 105 108 120 130 137 13 papers 21 28-29 31 37 67 78 87 101 103 105 108 120 found no difference in outcome based on time of admission, with five 26 33 84 100 130 concluding mixed outcomes.
Mortality rates and other outcome differences for patients admitted out of hours compared to in hours are for non-disease specific (unselected) hospital admissions are summarised in Tables 3–8.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 11 Medical Of the 20 papers analysing non-specific medical patient cohorts, 1 papers reported a positive impact; three found no difference in outcome of patients admitted out of hours compared with admissions on other times of the week. The remaining six reported mixed outcomes. Seventeen papers analysed administrative datasets, two undertook clinical dataset analysis and one paper a systematic review. The characteristics and findings of the included reviews and primary studies are summarised in Tables 3–5.
Fourteen papers 4 6 8-12 16 22 24 107 117 124 139 reviewed were multi-site studies, all analysing administrative data sets. Of these multi-site papers, nine 6 8 10-12 16 107 139 found differences in outcomes, two 117 124 found no difference and three 4 22 24 confirmed mixed outcomes for patients admitted out of hours compared to those who are not.
The majority (9) 4 10-12 16 22 107 117 124 of the papers analysed nights and weekends together with seven 10-12 16 37 107 127 of these papers finding a difference in outcome, three 117 124 37 finding no difference and three 4 22 127 finding mixed outcomes. Four 6 9 24 141 papers defined out of hours as the weekend (Saturday & Sunday) and compared this with all other times.Three 6 9 141 of these papers confirmed a difference in outcome for patients admitted out of hours and one 24 finding a mixed result. Two 98 139 papers compared weekends and nights separately finding a difference in outcome . One paper 147 compared each day to the next finding mixed outcomes. The remaining two papers did not clearly define the timeframe variable.
Of the papers assessing in-hospital mortality, five 6 12 16 98 107 found a difference in outcome; two 37 124 finding no difference and three studies 4 22 147 finding mixed outcomes. All of these studies analysed administrative data sets with only two 37 98 papers examining clinical data. Thirty-day mortality was examined in six papers with three finding a difference, three mixed and one concluding no difference in outcome. Only one paper 37 evaluated LoS, finding no difference. A breakdown of these variables is summarised in Appendix 5. Surgical Of the nine papers analysing non-specific surgical patient cohorts, six papers reported a positive impact; two found no difference in outcome of patients admitted out of hours compared with admissions on other times of the week and one reported mixed outcomes. Seven studies analysed administrative datasets and with the remaining four analysed clinical data. The characteristics and findings of the included reviews and primary studies are summarised in Tables 6–8.
Thirty-day mortality was analysed on four 7 17 31-32 of the nine papers. Five 14 32 87 93 155 included a review of in-hospital mortality. Only two papers evaluated differences in LoS 87 93. A breakdown of these characteristics is summarised in Appendix 5.
Four of the papers 7 14 17 87 reviewed identified analysis of elective surgical patients. Three papers 7 14 17 analysed administrative data sets and all finding a positive impact on out of hours admission compared with admission at other times.The remaining paper 87 analysed a single-site and after adjustment found that patients undergoing non-emergency surgery procedures out of hours did not seem to have an increased risk for mortality or morbidity. They concluded that performing elective procedures at night may be a safe solution for daytime overcrowding of operating rooms. Intensive Care Twenty papers analysed patients admitted to the ICU. Four papers reported a positive impact; ten found no difference in outcome of patients admitted out of hours compared with admissions on other times of the week and six reported mixed outcomes. Seventeen papers analysed clinical data. One paper was a systematic review which included eight of the primary studies named below and marked with a ‘*’. The remaining two studies analysed an administrative dataset. The characteristics and findings of the included reviews are summarised in Tables 9–11.
The majority (11) 79 84-85 100-101 103 108 111 130 137 of papers consisted of multi-site studies of which nine 79 100-101 103 108 111 130 137 analysed clinical datasets. Four of the papers reviewed LoS with three 21 33 78 finding no differential outcome. All of the papers were retrospective observational studies except for three 21 84 100 that were prospective examinations of a clinical dataset and analysed weekends and nights separately compared to weekday office hours.
12 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project TABLE 3: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Medical patients
Author, Year, Study Design Patient number and Type Main Findings Country Aylin, 2010, Retrospective analysis 4,317,866 inpatient admissions via ED Difference in mortality, for 23 (adjusted) for the top 100 highest mortality conditions. Of a total of UK.6 of administrative data: (5% 215,054 deaths). All acute hospitals in 4,317,866 emergency admissions, found 215,054 in-hospital deaths with an overall CMR of 5.0% 2005–2006. England. (5.2% for all weekend admissions & 4.9% for all weekday admissions). The overall adjusted odds of death for all emergency admissions was 10% higher (OR: 1.10; 95% CI: 1.08-1.11) in those
Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project patients admitted at the weekend compared with patients admitted during a weekday (p<0.001). Barba, 2006, Retrospective analysis of 35,993 ED admissions to hospital. (single-site). The study suggests a higher risk of death for patients. (OR: 1.40 in-hospital mortality) Risk of Spain.98 clinical data: 1999–2003. death within the first 48 hours is higher for patients admitted on weekends than for patients admitted on a weekday. Concha, 2013, Retrospective analysis 3,381,962 (917,257 weekend) inpatient Patterns consistent with both patient & care factors (early risk pattern-majority of excess deaths Australia.8 of administrative data: admission via ED. Multi-centre (501 public & occurred in first 24h & 48h of admission) but not consistent across DRGs. Risk of death out of 2000–2007. private hospitals) (18,282 weekend deaths). hours (OR: 2.12 (adjusted)) in 16 of 430 diagnostic groups. Cram, 2004, Retrospective analysis of 641,860 inpatients admitted via ED for Being admitted to hospital on a weekend was associated with slightly higher mortality risk than USA.9 administrative data: 1998 50 diagnoses (41,702 deaths). Most Acute being admitted to hospital on a weekday. In-hospital mortality (OR: 1.03) for 3 of 50 diagnoses. Hospital survey. US hospitals. Risk of death out of hours was greater in major teaching hospitals than minor or no teaching hospitals. Freemantle, Retrospective analysis 14,217,640 ED admissions &/or already an Admission on weekend was associated with increase in rise of subsequent death. Risk of 2012, UK.10 of administrative data: inpatient on weekend days (187,337 in-hospital death out of hours (HR: 1.16 in-hospital mortality for Sunday v. Wednesday HR: 1.11 in-hospital 2009–2010. deaths). NHS hospitals England. mortality for Saturday v. Wednesday). But being in-hospital at the weekend was associated with reduced death. Freemantle, Retrospective analysis 14,818,374 ED admissions &/or already an Patients admitted at the weekend are more like to be in the highest category of risk of death 2015, UK.11 of administrative data: inpatient of weekend days (280,788 deaths). & face increased likelihood of death even when severity of illness is accounted for. Risk of 2013–2014. NHS hospitals in England. death out of hours (HR: 1.1 on Saturdays & 1.15 on Sundays: p= 0.001) for death within 30 of admission compared to Wednesdays. Handel, 2012, Retrospective analysis 5,271,327 ED admissions to hospital. NHS There was a significantly increased probability of death associated with a weekend emergency UK.12 of administrative data: hospitals in Scotland. admission compared with admission on a weekday (OR: 1.27; 95% CI: 1.26–1.28, p<0.0001). 1999–2009. Despite a general reduction in mortality over the last 11 years, there was still a significant excess mortality associated with weekend emergency admissions. Marco, 2010, Retrospective analysis of 429,880 admissions from the ED (single-site) to Higher mortality associated with out of hours admission than admission during the week. Risk Spain.141 clinical data: 2005. the hospital. of death out of (RR:1.15 in-hospital mortality). Differences in mortality persisted after adjustment for age, sex, & coexisting disorders (OR: 1.071; 95% CI: 1.046–1.097). Analysis of deaths within 2 days after admission showed larger relative differences in mortality between weekend & weekday admissions. Ricciardi, 2014, Retrospective analysis 48,253,968 patient discharges (26,038,921 Risk of death out of hours (RR: 1.15 in-hospital mortality). USA.16 of administrative data: non-elective inpatient admissions. (multi-centre). 2003–2008. Schilling, 2010, Retrospective cohort study 166,920 admissions to EDs (39 sites) for Difference in mortality (OR: 0.32: CI: 0.11–0.54) associated with out of hours admission than USA.107 of administrative data: diagnosis of AMI, heart failure, stroke, admission during the week. 2003–2006. pneumonia, hip fracture GI Bleeding). Vest-Hansen, Retrospective cohort study of 174,192 acute medical admissions to hospitals. Whilst admission rates decreased from office hours to weekend hours there was an observed 2015, administrative data: 2010. increase in mortality 30-day mortality was 5.1% (95% CI: 5.0–5.3) after admission during Netherlands.139 weekday office hours, 5.7% (95% CI: 5.5–6) after admission during weekday (out of hours),
13 6.4% (95% CI: 6.1–6.7) after admission during weekend night time hours. Researchers concluded differences in severity of illness as the proportion admitted to hospital. TABLE 4: Data extraction of studies reporting no impact from admission out of hours on outcomes – Medical patients
Author, Year, Patient number and Study Design Main Findings Country Type Mikulich, Retrospective 49,337 medical Patients admitted at the weekend had an approximate 11% increased 30-day 2010,UK.117 cohort analysis of admissions single-site, in-hospital mortality, compared with a weekday admission; although this was administrative Dublin, Ireland hospitals. not statistically significant either before or after risk adjustment (30-day in- data: 2002-2009. hospital mortality 9.9% v. 9.0% unadjusted OR: 1.05; 95% CI: 0.88=1.24). The authors pointed out that “admission at the weekend was not independently predictive in a risk model that included illness severity (age & biochemical markers) & co-morbidity”. Powell, 2013, Retrospective 114,611 ED admissions ED admissions with a principal diagnosis consistent with sepsis, & found that USA.124 cohort study of from 576 hospitals. the difference for overall inpatient mortality (in terms of the weekend) was not administrative significant (17.9% v. 17.5%, p= 0.08). data: 2008. Schmulewitz, Retrospective 3,244 admissions (938 No significant risk of death out of hours for any of the selected conditions (OR: 2005, UK.37 cohort study of weekend admissions) 0.5 to 1.65 across diagnostic groups) compared to traditional working hours. clinical data: for 6 predetermined 2001. diagnoses (COPD, CVA, PE, CAP, GI Bleed & ‘collapse’) (single-site).
TABLE 5: Data extraction of studies reporting mixed outcomes for patients admitted out of hours’ – Medical patients
Author, Year, Patient number and Study Design Main Findings Country Type Becker, 2008, Retrospective 922,074 elderly Medicare Positive correlation between weekend hospitalisation & mortality. Weekend USA.22 analysis of claim admissions patients experience a 0.38 percentage point (p= 0.001) increase in 1-year administrative with AMI. mortality. Patients admitted on weekend significantly less likely to receive data: 1989– primary intensive treatment associated with AMI within the first days of 1998. admission. Weekend admission with AMI leads to lower subsequent expenditure but higher 1-year mortality & higher rate of one major cardiac complication, readmission with CHF. Out of hours admissions lead to inappropriate reduction in intensive medical care. The effects on weekend admission do vary with patient health status: while weekend admission leads to delay in invasive treatments for all patients, sicker patients (defined by inpatient admission the previous year) are less likely to experience delays. Bell, 2001, Retrospective 3,789,917 ED patients Greater severity of illness among patients admitted to acute care hospitals Canada.4 analysis of (all (190 Ontario out of hours would still raise questions about the adequacy of medical care administrative hospitals). Selected & staffing patterns. Significant effect in 23 (adjusted) for the top 100 highest data: 1988– DG hypothesized to be mortality conditions (OR: AAA 1.28; AE 5.28; PE 1.25; AMI 1.02; ICH 1.01; 1997. susceptible (AAA, AE, Hip #. 0.95). PE) or non-susceptible * Index cases: ruptured abdominal aortic aneurysm, acute epiglottis, & pulmonary (AMI, ICH, Hip #) to embolism. differences in outcome * Controls: Myocardial Infarction, intra cerebral haemorrhage, Hip Fracture Clarke, 2010, Retrospective 30,522 COPD, 17,910 Study found a significant effect for acute myocardial infarction (adjusted RR: Australia.24 analysis of AMI, 4,183 acute hip 1.15; 95% CI: 1.03–1.26: p= 0.007). Two-day in-hospital mortality showed administrative fracture and 1,781 intra- similar results. There was no significant effect on adjusted 30-day in-hospital data: cerebral haemorrhage mortality for COPD (RR: 0.92; 95% CI: 0.81–1.04: p= 0.222), intra cerebral 2003–2007. admissions. Queensland haemorrhage (RR:1.01; 95% CI: 0.86–1.16: p= 0.935) or acute hip fracture hospitals. (RR: 0.78; 95% CI: 0.54–1.03, p= 0.13). De Cordova, Systematic 66 Studies. Focus on weekends. Most studies (9 of 12) did not find an important 2012, USA.25 Review. association between patients admitted at night & mortality. Maggs, 2010, Retrospective 15,594 adults admitted After adjusting for age & sex, we did not detect a significantly increased UK.147 cohort analysis of under care of physician overall mortality in patients admitted at the weekend compared with those administrative (single-site). admitted on weekdays. However, total mortality was increased in patients data: 2007– admitted on Mondays & at night compared with those admitted in the 2008. daytime, & in all out of hours periods taken together compared with normal working hours. Smith, 2013, Retrospective 20,072 admissions to Mortality of patients admitted out of hours not significantly higher at 7-days (OR: USA.127 cohort study of single-site hospital. 1.10; 95% CI: 0.92–1.31; p= 0.312) or at 30-days (OR 1.07; 95% CI: 0.94–1.21; administrative p= 0.322). By contrast, they found adjusted public holiday mortality in the all data: 2008– public holidays was 48% higher at 7-days (OR: 1.48; 95% CI: 1.12–1.95; p= 2010. 0.006) & 27% higher at 30-days (OR: 1.27; 95% CI: 1.02–1.57; p= 0.031).
14 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project TABLE 6: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Surgical patients
Author, Year, Study Design Patient number and Type Main Findings Country Aylin, 2013, Retrospective 4,133,346 elective surgical patients. The study suggests a higher risk of death for patients who have elective surgical procedures carried out later in the UK. 7 analysis of NHS hospitals England. working week & at the weekend. There were 27,582 deaths within 30-days after 4,133,346 inpatient admissions for administrative elective operating room procedures (overall CMR 6.7 per 1000). The number of weekday & weekend procedures data: 2008–2010. decreased over the three years (by 4.5% & 26.8%, respectively). The adjusted odds of death were 44% & 82%
Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project higher, respectively, if the procedures were carried out on Friday (OR: 1.44; 95% CI: 1.39–1.50) or a weekend (OR: 1.82; 95% CI: 1.71–1.94) compared with Monday. Mohammed, Retrospective 1,535.267 elective surgical patients Difference (OR: 1.32 for elective patients in-hospital mortality; OR 1.09 for emergency patients in-hospital mortality). 2012, UK. 14 cohort analysis of (0.54% deaths) 3,105,249 emergency administrative surgical patients (6.67% deaths) (328 data: 2008. acute hospitals from 221 NHS Trusts) 3,105,249 emergency patients (6.67% deaths) (328 acute hospitals from 221 NHS Trusts). Ricciardi, 2011, Retrospective 29,991,621 (6,842,030 weekend Difference OR: 1.10 (in-hospital mortality); 10.5% higher at weekends 15 out of 26 major diagnostic categories. USA. 15 analysis of admissions) patients with a non- Higher co-morbidity scores for weekend admissions. administrative elective admission of data: 2003–2007. 20 US community hospitals. Ruiz, 2016, Retrospective 3,922,091 (26,409 deaths) elective The adjusted odds of death remained higher for Friday (OR: 1.48; 95% CI: 1.42 to 1.54), Saturday (OR: 1.97, UK. 17 observational procedures in 163 NHS hospitals. 95% CI: 1.83–2.12) & Sunday (OR: 1.67; 95% CI: 1.50–1.85) after adjusting for consultant seniority & patient study using characteristics. administrative Interestingly, in the Australian hospitals, data indicated that there was no significant daily change in the elevated data: 2009–2011. risk of death within 30-days for emergency admissions over weekends. These hospitals actually accounted for *examination of same largest number of emergency admissions overall. dataset of Aylin 2013. 7 Tadisina,2015, A serial cross 50,346 plastic surgery patients. US For patients who had body contouring procedures their mortality rates to be higher on weekend admission (3.7%) v. USA.93 sectional study of hospitals. weekdays (0.5%). administrative data: 2000–2010. Zapf, 2015, Retrospective 80,861 same day surgeries (19,078 Differential outcomes for paediatric surgery; increased mortality for children undergoing urgent surgery during the USA.155 cohort study of occurred during the weekend). weekend rates of wound complications (OR: 1.29: 95% CI: 1.05–1.58; P<0.05), & urinary tract infection (OR: 1.39; administrative 95% CI: 1.05–1.85; p<.05). Patients undergoing appendectomy had greater rates of transfusion (OR: 1.43; 95% CI: data: 2007–2010. 1.09–1.87; p= 0.01), wound complications (OR: 1.32; 95% CI: 1.04–1.68; p<0.05), UTI (OR: 1.76; 95% CI: 1.17– 2.67; p< 0 .01), & pneumonia (OR: 1.41; 95% CI: 1.05–1.88; p<0.05). Patients undergoing cholecystectomy had a greater duration of stay (p= 0.001) & greater charges (p= 0.003). 15 16 TABLE 7: Data extraction of studies reporting no impact from admission out of hours on outcomes – Surgical patients Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project
Author, Year, Study Design Patient number and Type Main Findings Country Orman, 2012, Retrospective cohort analysis of 94,768 (4% death) liver transplant No impact of weekend admission on outcome (RR: 0.99 at 30-days). USA. 31 clinical data: 1987–2010. operations (multi-site). Turrentine, Retrospective cohort analysis of 10,426 operative procedures Patients undergoing procedures at night had a greater prevalence of serious preoperative comorbid 2010, USA. 87 clinical data: 5 years. performed. conditions. Procedure complexity as measured by relative value unit did not differ between groups, but LoS was longer after night procedures (7.8 days v. 4.3 days, p<0.0001). Patients undergoing non-emergent general & vascular surgery procedures at night in an academic medical centre do not seem to be at increased risk for postoperative morbidity or mortality. Performing non-emergent procedures at night seems to be a safe solution for daytime overcrowding of operating rooms.
TABLE 8: Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Surgical patients
Author, Year, Study Design Patient number and Type Main Findings Country Ruiz, 2015, Retrospective cohort analysis 2,982,570 admissions hospital Difference in mortality rates for patients admitted out of hours for emergency admissions to 11 UK hospitals Global of administrative data: records from 28 metropolitan (OR: 1.08; 95% CI: 1.04–1.13 on Sunday) 5 US hospitals (OR: 1.13; 95% CI: 1.04–1.24 on Sunday) & 6 (UK, USA, 2009–2011. teaching hospitals in UK, USA, Netherlands hospitals (OR: 1.13: 95% CI: 1.09–1.33 on Saturday). No effect to Australian hospitals. Effect Netherland, *Examination of same data set of Netherlands & Australia. for all elective patients showed higher adjusted odds of 30-day postoperative death with Friday effect for 6 32 Australia). Aylin (2013) 7 Netherlands hospitals).
TABLE 9: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Intensive Care Unit patients.
Author, Year, Study Design Patient number and Type Main Findings Country Barnett*, 2002, Retrospective cohort analysis 156,136 admissions from 38 ICUs on None of the diagnoses had lower mortality for out of hours admission compared to the rest of the week. USA. 85 of administrative data: 28 Ohio hospitals. In-hospital death were 9% higher (OR: 1.09; 1.04–1.15; p= 0.001) for weekend admissions (Sat–Sun) 1991–1997. than in patients admitted mid-week (Tues–Thurs). However adjusted odds of death were also higher for patients admitted on Monday (OR: 1.09) or Friday (OR: 1.08). Findings were generally similar in analysis by admission type (medical v. surgical), hospital teaching status, & illness severity. Bhonagiri, Retrospective cohort analysis 245,057 admissions to 41 Australian Difference in mortality rates. Patients admitted after hours had a 17% hospital mortality rate compared with 2011, of administrative data: ICUs (48% after-hours admissions & 14% of patients admitted in hours. Weekend admissions had a 20% hospital mortality rate compared with Australia.111 2000–2008. 20% weekend admissions). 14% on weekdays. (p<0.001), with SMRs of 0.95 (95% CI: 0.94–0.97) & 0.92 (95% CI: 0.92–0.93.) Kuijsten, 2015, Retrospective cohort analysis 149,894 patients admitted to the ICU Study showed an increase in the risk of hospital mortality for patients admitted during off-hours compared Netherlands. 137 clinical data: 2002–2008. (multi-site). with patients admitted during office hours (RR: 1.059), & an increase of hospital mortality risk for patients admitted during the weekend compared with patients admitted during the week (RR: 1.103). Neuraz, 2015, Retrospective observational 5,718 ICU inpatient stays (8 ICUs). The risk of death increased with an (OR: 3.5; 95% CI: 1.3–9.1) when the patient-to-nurse ratio was greater France. 79 cohort clinical data: 2013. than 2.5, & by (OR: 2.0; 95% CI: 1.3–3.2) when the patient-to-physician ratio exceeded 14. The highest ratios occurred more frequently during the weekend for nursing staff & during the night for physicians (OR: 3.08; 95% CI: 1.34–7.09) with a greater risk of CPR in off-hours & holiday times than day shifts. TABLE 10: Data extraction of studies reporting no impact from admission out of hours on outcomes – Intensive Care patients
Author, Year, Study Design Patient number and Type Main Findings Country Arabi, 2006, Prospective cohort tudy ICU 2,093 ICU patients (single-site) There was no significant difference in-hospital mortality rates among the 3 time periods (36%, 36%, & Saudi Arabia. 21 clinical data: 1999–2003. 31% admitted weekdays; 35% 37%, respectively, p= 0.90). weeknights; 34% weekends. Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project
Arslankoylu, Retrospective cohort study of 210 admissions to PICU the (single- No significant difference for overall mortality rates between weekend & weekday admissions (12.2% 2008, Turkey. 67 clinical data: 2005–2006 (6 site) for patients aged 1mth–18 v. 17.4%, p= 0.245), & daytime & evening admissions (11.3% v. 15.4%: p= 0.254). There was also no months). years. significant difference between different admission times for within 24 hours, 48 hours & 72 hours mortality rates. Hixson, 2005, Retrospective cohort study 5,968 PICU patients aged 0 days– No significant difference in mortality (p= 0.146), weekend discharge/death (p= 0.348), nor evening PICU USA. 28 PICU clinical 1996–2003. 21yrs. (single-site). admission (p= 0.711) showed a significant relationship with mortality controlling for other significant factors. Limiting the scope to the emergency admissions subset, neither weekend admission (p= 0.135), weekend discharge/death (p= 0.278), nor evening PICU admission (p= 0.867) were significant predictors of mortality. Laupland, Retrospective cohort study of 7,380 patients (multi-site). No difference in mortality between admissions in working hours v. out of working hours once adjusted for 2011, clinical dataset: 2006–2010. case mix. Canada.101 Lee*, 2008, Retrospective cohort study of 848 (327 night; 248 weekend) No significant difference in ICU & hospital mortality for day & time of ICU admission. Singapore. 29 clinical severe head injury admissions to specialised database: 1999–2006. neuroscience ICU. Luyt*, 2007, Retrospective cohort analysis 51,643 ICU patients (33 857 No significant difference in mortality. After adjustment, in-hospital mortality was not higher for off-hours France. 103 of prospectively collected admitted off-hours) from 23 sites. admissions than weekdays day admissions & even remained slightly lower (adjusted OR: 0.93; 95% CI: clinical data 2000–2003. 0.87–0.98). Meynaar*, Retrospective cohort study 6,725 ICU (2,712 (32.3%) weekend) No effect. Mortality ratios were similar for patients admitted during off-hours & patients admitted during 2009, with prospectively collected admissions across 3 hospitals. daytime unadjusted (OR: 1.36; CI: 1.20–1.55). Netherlands. 105 clinical ICU database: 2004–2007. Morales*, 2003, Retrospective cohort study 6,034 ICU admissions (single-site). No increase in mortality for night admissions: patients admitted at nights had lover morality rate (13.9 v. USA. 78 of clinical ICU database: 17.2; p <0.001). 1995–2000. Numa, 2006, Retrospective cohort analysis 6,980 PICU admissions (4,456 non- Patients admitted after hours had a lower risk adjusted mortality than those admitted during normal Australia.120 of prospectively collected elective) to 12 bed PICU Sydney, working hours, with death (OR: 0.712; 95% CI: 0.518–0.980: p= 0.037). clinical PICU database: Australia. 1997–2006. Wunsch*, Retrospective cohort study 56,250 ICU patients (102 sites). Higher crude mortality for weekend & night admissions. (Fri OR: 1.19 1.11-1.27; Sat OR: 1.41 2004 UK.108 clinical ICU database: 1.32–1.52 or Sun OR: 1.56, 1.45–1.68). No differential outcome after appropriate adjustment for case mix 1995–2000. day of the week & time of day of ICU admission were not associated with significant differences in-hospital mortality (Sat OR: 1.03 0.9–1.12; Sun OR: 1.09 1.00–1.19). 17 TABLE 11: Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Intensive Care patients
Author, Year, Study Design Patient number and Type Main Findings Country Arias,2004, Retrospective 20,547 admissions to 15 PICU’s. There was no association between mortality rates & the day of USA. 130 cohort study admission (weekend admissions v. weekday admissions). Paediatric of clinical patients admitted to the PICU during evening hours had higher odds ICU database: of death (OR: 1.28; 95% CI: 1.00–1.62) than did those admitted 1995–2001. during daytime hours. Subgroup analyses revealed higher odds of death among patients admitted with shock (OR: 4.09; 95% CI: 1.65–10.1), with congenital cardiovascular disease (OR: 3.90; 95% CI: 1.37–11.1), or after cardiac arrest (OR: 1.80; 95% CI: 1.04–3.13). Cavallazzi, Systematic 10 cohort studies (8 included Mixed results. Night-time admissions was not associated with an 2010, Review night-time admission, 6 evaluated increased mortality (OR: 1; 95% CI: 0.87–1.17; p= 0.956); however, 23 Global. (studies included weekend admissions). patients admitted over the weekend had a significant increase in the marked *) adjusted risk of death (OR: 1.08; CI: 95% 1.04–1.13 p<0.001). Ensminger*, Retrospective 29,084 ICU patients (single-site). Higher crude mortality for out of hours admissions. The increased 2004, USA. cohort study mortality persisted after risk adjustment for surgical patients but not 26 of clinical medical patients (OR: 1.23 in-hospital mortality). No significant effect ICU database: for medical or multispecialty patients with weekend ICU admission 1994–2002. not found to be independently associated with increased hospital mortality (OR: 1.057; 95% CI: 0.952–1.74). Laupland*, Prospective 20,466 patients (24 204 ICU Mixed results. No effect after controlling for confounding variables 2008, cohort study admissions (9,987 weekend weekend admission. Weekend admission was not associated with Canada. 100 of clinical admissions) (multi-centre). death. (OR: 1.05; CI: 95% 0.95–1.17; p = 0.328). However, night ICU database: admission was independently associated with mortality (OR: 1.37; CI: 2002–2006. 95% 1.24–1.50; p<0.001). clinical data included & used to adjust for severity of illness – no difference was found after this adjustment. Sheu*, Retrospective 611 ICU patients (single-site) 372 The differences were not significant, the adjusted odds of hospital 2007, cohort study patients admitted off-hours. death were higher for patients admitted to the ICU at night, compared Taiwan. 33 of clinical to those admitted during the days (OR: 1.46; 95% CI: 0.99–2.14; p = ICU database: 0.056). On the other hand, the adjusted odds of hospital death were 2006. not higher for patients admitted to ICU on weekends, compared with those admitted to ICU on weekdays (OR: 0.79; 95% CI: 0.5–1.25; P= 0.31). Uusaro*, Prospective 23,134 emergency admissions Risk of dying in the ICU was higher during ‘out of office hours’ 2003, cohort study (4,677 [20.2% weekend admissions] as compared with ‘office hours’. Weekend ICU admissions are Finland. 84 Multiple centre across 18 ICUs); 126,754 acute associated with increased ICU mortality (by 20%) in critically ill ICU clinical care admissions from 190 patients. (OR: 1.20; 95% CI: 1.01–1.43). However, admissions in the database. emergency departments. evening & during the night-time were not associated with increased risk of death.
Disease specific (selected) patients Cardio Respiratory We identified 26 relevant studies from database searches and snowballing articles. Ten studies confirmed an increased risk of mortality for patients admitted out of hours compared to traditional working hours. Eleven found no difference in outcome with five concluding mixed outcomes. Fifteen studies analysed clinical datasets with 1 analysing administrative datasets. Only one study was a prospective cohort investigation. The characteristics and findings of the included reviews and primary studies are summarised in Tables 12–14.
An additional 11 studies 4 6 9 10 12 22 24 37 107 139 141 are not included in the below summary table, as they were examined in Tables 3–5. Only one 37 of these studies analysed clinical data, finding no difference in outcome. The remaining studies analysed administrative data finding a difference in outcome for patients admitted out of hours apart from three papers 4 22 24 which concluded mixed outcomes.
In large administrative datasets AMI, AF, PE, PCI and pneumonia were associated with poorer outcomes for patients that are admitted to hospital out of hours. The majority of studies (15) investigated AMI (incl. NSTEMI, STEMI). 34 44 50 59 62 65-66 69 83 110 112 113 115-116 128 Ten of the studies analysed clinical data sets, with seven 59 62 65 83 110 136 158 finding no differences in outcome while the remaining three 34 112 115 found the outcomes to be mixed. Four papers investigated the outcomes for patients undergoing PCI 52 63 65 82 by analysing clinical data sets with two finding 63 65 no difference.
18 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project Two studies investigated patients admitted with COPD 131 153 with the administrative dataset concluding unequal outcomes for patients admitted out of hours and mixed results for studies analysing clinical datasets.
Nineteen studies analysed multi-centre, with nine 73 44 50 69 80 82 116 152 153 concluding an increased risk of mortality. Eight 44 50 69 73 80 116 152 153 of these analysed administrative datasets. Eight multi-site studies found no difference in outcome, with six 59 62 110 128 136 145 analysing clinical datasets. The remaining two 115 131 papers interrogating clinical data found the outcomes mixed. Seventeen 34 44 52 59 63 65 66 69 80 82 83 92 110 115 116 131 153 of these multisite studies investigated in-hospital mortality and seven 50 62 65 66 73 113 128 136 152 investigated mortality at 30-days. Two 73 92 papers investigated whether LoS was increased for out of hours admission with one 73 paper finding an increased LoS. Eight papers investigated treatment delay with the majority (5) 59 62 63 65 66 finding no difference in delay in treatment for patients admitted out of hours.
TABLE 12: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Cardio Respiratory patients
Author, Year, Patient Number and Study design Main Findings Country Type Aujesky, 2009, Retrospective cohort 15,531 acute PE Differential outcome. Patients with PE who are admitted on USA. 73 analysis of administrative discharges (3,286 weekends have a significantly higher short-term mortality than data: 2000–2002. weekend admissions) patients admitted on weekdays. (unadjusted) OR: 1.17; 95% CI: from 186 hospitals. 1.03–1.34). Chang, 2012. Retrospective cohort 788,011 pneumonia Patients admitted on weekends had 3% higher odds of 30-day Taiwan. 152 analysis of administrative admissions. death compared with those admitted on weekdays (OR: 1.03; National Health Insurance 95% CI: 1.01–1.05). claims: 1997–2008. Deshmukh, Retrospective cohort 86,497 AF hospitalisations Patients admitted on weekends experienced greater 2012, USA. 80 analysis of administrative (16,949 weekends) with proportions of in-hospital mortality than those admitted data: 2008. AF. on weekdays (1.1% v. .9%; p= 0.01). The use of a cardio- vascular procedure for AF on weekends was lower than that on a weekday (7.9% v. 16.2%: p<0.0001; OR: 0.5; 95% CI: 0.45–0.55; p<0.0001). After adjusting for patient & hospital characteristics & disease severity, the adjusted in-hospital mortality odds were greater for weekend admissions (OR: 1.23; 95% CI: 1.03–1.51; p<0.0001). Glasser, 2008, Retrospective cohort study Of 10,948, 685 patients Patients presenting in off-hours were more likely to present USA. 82 of clinical: 1997–2006. underwent PCI (228 with greater severity of illness.Patients undergoing PCI on procedures during off- weekends had better outcomes during daytime than night time. hours). Gyenes, 2013, Retrospective cohort 6,711 NSTEMI patients Patients admitted on weekends had higher adjusted mortality & Canada. 44 analysis of administrative (19,56 admitted at the cardiovascular event rates. Higher risk patients were less likely data: 1999–2003. weekend). Canadian to undergo angiography & waited longer, with higher observed hospitals. in-hospital event rates. Weekend admission was independently associated with higher mortality (OR: 1.52; 95% CI: 1.15–2.01; p= 0.004). Khera, 2013, Retrospective cohort 1,434,579 patients with a Out of hours admission & multi-vessel PCI were independent USA. 116 analysis of administrative STEMI. US hospitals. predictors of in-hospital mortality among patients who data: 2001–2010. underwent PCI for STEMI. Kostis, 2007, Retrospective cohort 231,164 heart attack Significant difference of mortality (HR: 1.048). 30-day mortality USA. 50 analysis of administrative patients from all New after adjustment for patient & disease characteristics. Less data: 1987–2002. Jersey hospitals. frequent use of invasive cardiac procedures - not significant (HR: 1.023) after further adjustments for cardiac procedures. Kruth, 2008, Cohort study of 45 508 consecutive The study found a significant higher in-hospital mortality (11.1 v. Germany. 69 administrative data: AMI patients from 385 9.4%: p= 0.01) & at night there was a trend to higher in-hospital 1994–2002. German hospitals with mortality when compared with regular working hours (10.6 v. & without onsite cardiac 9.4%, p= 0.07). catheterization facilities. Lairez, 2009, Retrospective cohort 2,266 consecutive Mortality higher for night-time & weekend group. This was France. 52 analysis of prospectively emergency PCIs (870 sustained after adjusting for risk factors. collected clinical data: during off-hours). 2005–2008. Suissa, 2014, Cohort study of 323,895 patients with Found mortality was higher for weekend (OR: 1.06; 95% CI: Canada. 153 administrative data: COPD & pneumonia. 1.03–1.09) but not Friday admissions (OR: 95%; CI: 0.65–0.99). 1990–2007.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 19 20 TABLE 13: Data extraction of studies reporting no impact from admission out of hours on outcomes – Cardio Respiratory patients Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project
Author, Year, Study Design Patient number and Type Main Findings Country Al-lawati, 2012, Prospective cohort analysis of 4,616 AMI patients with ACS (76% on No significant effect in mortality in 1-month (OR: 0.88; 95% CI: 0.68–1.14) & 1-year mortality UAE. 128 clinical data: 2008–2009. weekdays). (OR: 0.88; 95% CI: 0.70–1.10) between weekday & weekend admissions. Similarly, there were no significant differences in 1-month (OR: 0.92; 95% CI: 0.73–1.15) & 1-year mortality (OR: 0.98; 95% CI: 0.80–1.20), between nights & day admissions. Berger, 2008, Retrospective cohort analysis 12,480 AMI patients (48% admitted out of No differences in terms of in-hospital survival rates between the 2 groups (91.5% v. 91.2%; Switzerland. 110 of clinical data: 1997–2006. hours) 106 hospitals, Switzerland. p= 0.633) or MACE-free survival rates (both 88.5%; p= 1.000) were noted. In conclusion, the outcome of patients with AMI admitted out of hours was the same compared with those with a weekday admission. Of predictors for in-hospital outcome, timing of admission had no significant influence on mortality &/or MACE incidence. Fonarow, 2008, Retrospective cohort analysis 48,612 patient data from 259 US hospitals When analysed by admission of weekend v. weekday the in-hospital death rate did not USA. 145 of clinical data: 2003–2004. with heart failure. significantly differ. Weekend admission was not associated with increase for in-hospital rate compared with weekday admission (OR: 0.99; 95% CI: 0.84–1.17). Differences in out of hours admission was not associated with a significant increase in post discharge death, death/ rehospitalisation, or rehospitalisation rates on multivariable analysis (HR: 1.16, 95% CI: 0.87– 1.55; OR: 1.03; 95% CI: 0.91–1.18; & OR: 1.07; 95% CI: 0.91–1.26, respectively). Jneid,2008, Retrospective cohort analysis 62,814 (982 out of hours admissions). No measurable differences, however, were found in in-hospital mortality between regular hours USA. 59 of the GWGCAD clinical & off-hours in the overall AMI, ST-elevated MI, & non-ST-elevated MI cohorts (adjusted OR: database: 2000–2005. 0.99; 95% CI: 0.93–1.06; adjusted OR: 1.05; 95% CI: 0.94–1.18; & adjusted OR: 0.97; 95% CI: 0.90–1.04, respectively). Koike, 2011, Retrospective cohort analysis 173,137 cases of out-of-hospital cardiac No significant differences were found between weekday & weekend/holiday cases, with ORs Japan. 136 of clinical data: 2005–2008. arrest Japanese hospitals. of 1.00 (95% CI: 0.96–1.04; p= 0.96) for 1-month survival & 0.99 (95% CI: 0.94–1.04; p= 0.78) for neurologically favourable 1-month survival. Even after adjusting for confounding factors, admission day (weekday v. weekend/holiday) had no effect on 1-month survival or neurologically favourable 1-month survival. In contrast, daytime admission was associated with significantly better outcomes than night time admissions. Matsui, 2007, Retrospective cohort analysis 4,805 consecutive patients with admitted There were no obvious differences in outcome for Japanese AMI patients in the weekday or Japan. 62 of clinical data: 2000–2003. within 48hours of onset AMI (1,276 weekend weekend onset group, suggesting the quality of the Japanese healthcare system is similar for the onset). entire week. Norman, 2012, Retrospectively analysis 2,571 PCI treated STEMI patients (single- There was no difference in mortality between weekday & weekend groups (OR: 1.09; 95% CI: UK. 63 of prospectively collected site). 0.82–1.46; p= 0.57). Similarly, no increase in mortality was seen in patients who underwent PCI clinical data: 2008–2011. at night (22:00–06:00 hours). Orandi, 2014, Retrospectively analysis 63,768 patients with an ischaemic lower limb. No statistically significant association between weekend admission & in-hospital mortality (OR: USA. 92 of administrative data: US hospitals. 1.15; 95% CI: 1.06–1.25; p= 0.10). 2005–2010. Rahthod, 2013, Retrospective analysis of 3,347 STEMI patients UK Hospitals. In-hospital mortality rates were comparable between the weekday & weekend groups (3.6% v. UK. 66 administrative: 2004–2012. 3.2%) with day of presentation not predictive of outcome (OR: 1.25; 95% CI: 0.74–2.11). Showkathali, Retrospective analysis of 1,471 STEMI patients (single-site). Large Found a similar result. In-hospital mortality, 30-day mortality, & 1-year mortality was not different 2013, UK. 65 clinical data: 2009–2011. cardiothoracic centre in the UK with 24.7 between weekday & weekend groups. primary PCI service. Sorita,2014 Retrospective analysis of 3,422 AMI patients admitted off-hours (2,664 Patients who were admitted during off-hours did not have higher mortality or readmission rates USA. 83 clinical data: 1998–2010. admitted in hours) at a single medical centre. as compared with one admitted during regular hours. TABLE 14: Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Cardio Respiratory patients
Author, Year, Study Design Patient number and Type Main Findings Country Brims, 2011, Retrospective analysis of 9,915 acute admissions The adjusted OR for death on day 1 after winter weekend UK.131 clinical data: 1997–2004. with exacerbation of admission was 2.89 (95% CI: 1.04–8.08). But after opening COPD. a MAU, death day 1 after weekend admission fell from OR: 3.63; 95% CI: 1.15–11.5 to 1.65; 95% CI: 0.14–19.01. Hong, 2010, Retrospective analysis 97,466 heart attack Differences in the case fatality rate of AMI patients admitted South Korea. of Korea National patients. on weekdays & on weekends in Korea are caused by 113 Health Insurance Claims differences in the rate of performance of medical or invasive (administrative) data: procedures. 30-day fatality rate was insignificantly different 2003–2007. after adjustment for medical or invasive management (OR: 1.05; 95% CI: 0.99–1.11). Hansen, 2013 Retrospective analysis of 92,164 AMI patients Mortality rates were higher on weekends within seven days of Netherlands.112 clinical data: 1997–2009. admitted to a admission in 1997–99 (absolute difference ranging from 0.8– Danish hospital. 1.1%), with weekend–weekday HR: 1.13 (95% CI: 1.03–1.23) at day HR: 2 & 1.10 (95% CI: 1.01–1.18) at day 7. But there were no significant differences in 2000–09 & suggesting an attenuation of the initial ‘weekend-effect’, perhaps relating to more equitable care. Keatinge, Retrospective analysis of NHS hospitals south east No adverse effect on mortality was apparent within 2-days 2005, UK.115 clinical data: 1989–2001. England. from reduction in medical services at weekends. However, respiratory deaths accelerated sharply after reduction in elective & emergency admissions at Christmas, when rates of infection & mortality from respiratory disease were high. Magid, 2008, Retrospective analysis of 68,439 patients with Differences in mortality for STEMI (OR: 1.07). Increased LoS USA.34 clinical data: 1999–2002. ST segment elevation associated with PCI but not fibrinolytic therapy. No difference myocardial infarction in mortality out of hours were identified when they looked at treatment with fibrinolytic two separate groups. therapy & 33,647 treated with percutaneous coronary intervention.
Gastroenterology We identified15 studies from the database searches investigating outcomes of patients admitted to hospital with gastroenterological conditions out of hours. six papers reported differences in outcome for patients admitted out of hours compared to those who were not, eight found no difference and the remaining one paper reported a mixed outcome. Eight of the papers analysed administrative datasets, eight clinical datasets and one paper a meta-analysis. 45 Ten of the studies reviewed are also included in the meta-analysis and are marked in the tables below with an ‘*’. The characteristics and findings of the included reviews and primary studies are summarised in Tables 15–17.
An additional ten 4 6 9-12 14 37 127 139 studies are not included in the below summary as they are examined in detail above within Tables 3–5. Only one paper 37 analysed clinical data and the only paper concluding no difference in outcome. The remaining studies all analysed administrative data finding six 9-12 14 139 papers concluding difference in outcome for patients admitted out of hours and three 4 6 127 papers found the outcomes mixed
The majority of studies (8) related to UGIB, 27 38 42 43 49 54 74 89 with others specifically analysing UGIB from peptic ulcer, 54 109 diverticulitis, 88 cirrhosis, 41 appendicitis 35 and oesophageal variceal haemorrhage. 91
Eight papers 27 35 38 41 49 89 91 109 found that patients admitted to hospital out of hours had similar adjusted in-hospital rates. All of these papers analysed clinical data with one 49 utilised both clinical and administrative data in a prospective cohort study.
Twelve studies analysed data from multiple sites. 35 38 39 42 43 49 54 74 88 89 91 109 Of these, the papers finding no difference in outcome, three 38 89 49 interrogated administrative data with the remaining three 35 91 109 using clinical data. Of the five multi-site studies finding a difference in outcome four 43 54 74 88 analysed administrative data with one paper 42 investigating a clinical dataset.
All seven 35 38–39 41 45 49 54 papers investigating a risk of delay in treatment for patients admitted out of hours found one. However only one paper reported this risk impacting on patients’ LoS. 54 A breakdown of these variables are summarised in Appendix 4 & 5.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 21 TABLE 15: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Gastroenterology patients
Author, Year, Patient Number Study Design Main Findings Country and Type Button*, 2011, Retrospective cohort 24,421 admissions for UGIB Differential outcome. Adjusted cast fatality for NVUGIB was higher UK. 74 administrative study: among 22,299 from hospital for weekend (OR: 1.13; CI: 1.02–1.25) & public holiday (OR: 1.48; 1999–2007. inpatient & mortality data. CI: 1.10–1.98) admissions. UGIB case fatality was 13% higher for we admissions & 41% higher for admissions on public holidays. Not explained by case mix. DeGroot*, Prospective cohort clinical 571 patients presenting with Patients admitted during the weekend had higher mortality rate 2012, study: 2009–2011. UGIB to emergency from 8 than patients admitted during the week (9% v. 3%; adjusted Netherlands. 42 hospitals. OR: 2.68; 95% CI: 1.07–6.72). Weekend admissions were not associated with other adverse outcomes. Patients admitted during the evening had a significantly longer time to endoscopy (15, 22 &16 hours for day, evening & night admissions respectively, p<0.01). This may be due to the fact that these patients have more severe outcomes. Dorn*,2010, Retrospective analysis Multi-site study of 98,975 Out of hours admission for UGIB is associated with an increased USA. 43 administrative data: patients admitted to risk of death (HR: 1.09; CI: 1.00–1.18), slightly longer LoS, & 1998–2003. hospitals (23,339 on marginally higher inpatient charges. Discrepancies in the use & weekend) across 31 states timing of endoscopy do not account for differences. with UGIB. US hospitals. Hinds,2014, Meta-Analysis. 11 studies (10 examined Patients admitted with UGIB on weekend exhibited a statistically 45 Global. (studies included mortality); 6 examined need significant increase in mortality (OR: 1.13; 95% CI: 1.06–1.20; marked *) for surgery; 6 examined time p<0.01), need for surgery (OR: 2.46; 95% CI: 1.51–3.99; p<0.01), to endoscopy; 4 evaluated & time to endoscopy (MD 2.68; 95% CI: 0.017–5.20; p= 0.04) as endoscopy of admission day compared to patient with UGIB on a weekday. Patients with UGIB & 4 evaluated hospital LoS). admitted on weekend experienced statistically significant less endoscopy on day of admission (OR: 0.72; 95% CI: 0.62–0.85). No difference was noted between the two for LoS. Shaheen*, Retrospective cohort study 237,412 admissions to Higher mortality overall (13% relative increase in risk) was noted in 2009, of administrative data: 3,166 hospitals for peptic patients admitted to urban teaching hospitals (OR: 1.16; 95% CI: Canada. 54 1993–2005. ulcer related UGIB. 1.06–1.26). Patients admitted to hospital at the weekend for peptic ulcer-related haemorrhage have higher mortality & more frequently undergo surgery. Worni, 2012, Retrospective cohort study 31,832 admissions for acute Significant differential outcome for postoperative complications USA. 88 of administrative data: diverticulitis. (OR: 1.1) & non-routine hospital discharge OR: 1.33) compared 2002–2008. with weekday admission.
TABLE 16: Data extraction of studies reporting no impact from admission out of hours on outcomes – Gastroenterology patients
Author, Year, Study Design Patient number and Type Main Findings Country Abougergi, Retrospective cohort 202,340 NVUGIB. US Found that patients with NVUGIB admitted on weekends had 2014, USA. 38 analysis of administrative hospitals. similar adjusted in-hospital mortality rates (OR: 1.11; 95% CI: data: 2009. 0.93–1.30). Ahmed, Retrospective cohort 60,643 patients with UGIB. There was a significant reduction in 30-day case fatality from 2015, UK. 89 analysis of administrative NHS Scotland. 10.3%–8.8% (p<0.001) over 10 years. Patients admitted with data: 2000–2010. UGIB at weekends had a higher 30-day case fatality compared with those admitted on weekdays (p<0.001) after adjusting for comorbidities. Byun*, 2012, Retrospective cohort 294 patients with No differential outcome was identified. 23% of 74 patients with a South Korea. analysis of clinical data: cirrhosis admitted for the weekend admission & 21% of 220 within a weekday admission 41 2005–2009. management of UGIB. died during hospitalisation (p= 0.872). Haas*, 2010, Retrospective analysis of 174 UGIB admissions No differential outcome. Adverse outcomes were not associated USA. 27 clinical data: 2008. patients (50 weekend) with weekend admission (weekend: 36 of 50 [72%]; weekday: 84 (single-site). of 124 [68%] p= 0.583). Jairaith*, Large prospective 6,749 UGIB patients No differential outcome was found in UGIB. After adjustment for 2011, UK. 49 cohort study clinical & admitted to 212 hospitals. confounders there were no evidence of a difference between administrative data: weekend & weekday mortality (OR: 0.93; 95% CI: 0.75–1.16). 2007. (Adjusted for clinical severity) Myers*, 2009, Retrospective analysis of 36,734 admissions (EVH) After adjusting for confounding factors, including time of Canada. 91 volunteer clinical stroke (2,207 hospitals). endoscopy. No impact of weekend admission or outcome (OR: database: 1998–2005. 1.05; 95% CI: 0.97–1.14).
22 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project TABLE 16: Data extraction of studies reporting no impact from admission out of hours on outcomes – Gastroenterology patients
Author, Year, Study Design Patient number and Type Main Findings Country Worni, 2012(b), Retrospective cohort study 151,774 laparoscopic No impact of weekend admission on outcome (OR: 1.37). USA. 35 of clinical data: 1999–2008. appendectomy for acute appendicitis admissions. Youn*, 2012, Retrospective cohort study 388 UGIB patients (due to Most patients (97%) had undergone early endoscopy but South Korea. 109 of clinical data: 2007–2009. peptic ulcer) from 4 referral the mortality rate was not different between the two groups centres. (1.8% overall v. 1.6% on weekend).
TABLE 17: Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Gastroenterology patients
Author, Year, Study Design Patient number and Type Main Findings Country Ananthakrishnan*, Retrospective cohort (cross 28,820 discharges with AVH Out of hours admission was not predictive of in-hospital 2009, USA. 39 sectional) analysis of & 391,119 discharges with mortality in patients with AVH (OR: 0.94; 95% CI: 0.75– administrative data 2004. acute (NVUGIB). 1.18), but was associated with lower likelihood of early endoscopy in nonteaching hospitals (OR: 0.65; 95%; CI: 0.51–0.82). Higher mortality among NVUGIB patients admitted on weekends (OR: 1.21; 95% CI: 1.09–1.35).
Haematology-oncology and other medical specialities From the database searches, four studies investigated the outcome for patients admitted to hospital with haematology-oncology and other medical specialities. All papers reported no difference in outcome for patients admitted out of hours compared with admissions on other times of the week. Two of the papers analysed clinical data set with the other two investigating an administrative dataset. The characteristics and findings of the included reviews and primary studies are summarised below in Table 18.
A further 12 4 6 9-12 16 107 124 127 139 141 investigated a number of cancer conditions which out of hours admission was associated with a statistical significant risk of mortality for patients admitted out of hours.The characteristics and findings of these papers are summarised in Tables 3–5.
TABLE 18: Data extraction of studies reporting no impact from admission out of hours on outcomes – Haematology-oncology and other medical specialities
Author, Year, Study Design Patient number and Type Main Findings Country Bejanyan, Retrospective follow up 422 acute myeloid leukaemia Weekend admissions had lower early mortality (p= 0.04) 2010, USA. 40 cohort study of clinical (103 weekends) admissions. & 30-day mortality (p= 0.02). In multivariate analysis, only data: 1994–2008. time to TLC remained significantly longer for weekend admissions (p<0.001). Goodman, Retrospective cohort study 12,073 patients with acute Found that those admitted on a weekend did not have 2014, USA. 47 of clinical data: 1999–2011. leukaemia (2009 patients an increased mortality (OR: 1.0; 95% CI: 0.8–1.6). While admitted at the weekend). paediatric patients with newly diagnosed leukaemia admitted on weekends do not have higher mortality rates, they have a prolonged LoS, increased time to chemotherapy, & higher risk for respiratory failure. Patients who are severely ill at presentation represent a higher proportion of weekend index admissions. Laupland, 2010, Retrospective cohort 7,722 patients presenting with No impact of weekend admission or outcome (OR: 0.99: Canada. 138 analysis of administrative community onset bloodstream 95% CI: 0.83–1.16). data: 2000–2008. infections (4 sites Calgary). Schimd,2014, Retrospective cohort study 534,011 patients with No difference in mortality was found. In multivariate USA. 142 of administrative data: metastatic carcinoma of the analysis, out of hours admission was associated with an 1998–2009. prostate. increased likelihood of complications (OR: 1.15; 95% CI: 1.11–1.19) & mortality (OR: 1.20; 95% CI:1.14–1.27).
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 23 Nephrology and Renal disease One study was identified from the database searches which observed a higher risk of death among patients withAKI admitted at the weekend. The characteristics and findings of the included review of this study is summarised in Table 19. A further ten 6 8 9 10 11 12 141 16 107 139 papers investigated a the outcomes of patients admitted with various renal conditions for patients admitted out of hours. The characteristics and findings of these papers are summarised in the non-disease specific (unselected) medical patients section.
TABLE 19: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Nephrology and renal transplant patients
Author, Year, Study Design Patient number and Type Main Findings Country James, 2010, Retrospective cohort study 963,730 with diagnosis of Compared with admission on a weekday, patients admitted USA. 13 of administrative (NIS AKI, US hospitals. 214,962 with a primary diagnosis of AKI on a weekend had higher database): 2003–2006. admissions (22%) designated odds of death [adjusted OR: 1.07; 95% CI: 1.02–1.12]. The AKI as the primary reason risk for death with admission on a weekend for AKI was for admission (45,203 on a more pronounced in smaller hospitals (adjusted OR: 1.17; weekend & 169,759 on a 95% CI: 1.03–1.33) compared with larger hospitals (adjusted weekday). OR: 1.07; 95% CI: 1.01–1.13). Increased mortality was also associated with weekend admission among patients with AKI as a secondary diagnosis across a spectrum of co- existing medical diagnoses. In conclusion, among patients hospitalised with AKI, weekend admission is associated with a higher risk for death compared with admission on a weekday.
Neuroscience Along with cardio respiratory and gastroenterology patients, the most well studied disease is stroke patients, with varying results regarding the outcome of patients admitted out of hours.
Thirty relevant papers were identified investigating risk of mortality for patients admitted out of hours with neurological (including neurosurgical) conditions. Fourteen papers reported a greater risk of mortality for patients admitted out of hours compared with admissions at other times, seven papers found no difference in risk with nine reporting mixed outcomes. Nineteen papers interrogated administrative datasets with the remaining 11 investigating clinical datasets. The characteristics and findings of the included reviews and primary studies are summarised in Tables 20–22.
Another 13 papers 4 6 9-12 16 22 93 107 124 127 139 are not included in the below summary table, as they were examined in Tables 3–5.
Two studies 134 151 were identified as neurosurgical, of which both found an increased risk of death (30-days) for out of hours admission.
We identified five papers which were prospective cohort analysis studies. Three 48 68 119 studies found an increased risk of mortality for patients admitted out of hours all of which analysed clinical data; one 19 paper concluded no risk in mortality from analysis of administrative data set; the one 55 paper that found mixed outcomes interrogated clinical data. One paper 36 conducted a post hoc analysis of an international multi-centre RCT finding no difference in outcome form patients.
Three 46 48 53 of eight 19 20 46 48 53 55 58 60 papers investigating treatment delay found increased delay for patient admitted out of hours. Of the studies (3) 19 20 90 comparing LoS all three did not find a difference in LoS for patients admitted out of hours compare to those admitted during traditional working hours. A breakdown of these variables are summarised in Appendix 5.
The majority (20) of papers investigated nights and weekends together with 15 investigating administrative datasets and 5 clinical datasets. Of the papers interrogating administrative datasets, seven 46 53 76 122 125 126 159 concluded an increased risk in mortality for patients admitted out of hours. Five 58 60 71 81 90 found mixed outcomes and three 19 20 75 papers found no difference. Of the clinical data set studies, two 68 119 found and difference in outcomes, two 55 121 found mixed and one 114 found no difference in risk in mortality. Of the papers (4) comparing the outcomes for patients admitted day time, night time and weekends separately, two 48 134 found an increased risk for patients admitted at nights and or weekends and two 36 129 did not find any difference in outcomes for patients admitted nights and or weekends compared to day-time hours. Only two papers investigated weekends compared to all other times in the week with one 144 finding a mixed outcome and the other 143 concluding an increased risk of mortality for patients admitted during these times. A breakdown of these variables is summarised in Appendix 6.
Of the papers which demonstrated excess mortality associated with out of hours admission for acute stroke, the effect size has generally been modest and not all studies have demonstrated such an effect, particularly after adjustment for clinical case mix and admission delay.
24 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project TABLE 20: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Neuro Surgery patients
Author, Year, Study Design Patient number and Type Main Findings Country Busl , 2013, Retrospective cohort study 14,093 patients with acute non- One in 9 patients with non-traumatic SDH dies during USA 151 of administrative data: traumatic SDH. US hospitals. hospitalisation. Among the several predictors of in-hospital 1995–2001. mortality, the differential outcome & treatment with surgical evacuation are potentially modifiable factors. Further investigation may lead to improvements in management & outcomes. Multivariate analysis, weekend admission (OR: 1.19; 95% CI: 1.02–1.38) was an independent predictor of in-hospital mortality. Desai, Retrospective analysis of 580 children undergoing After multivariate analysis, children undergoing procedures 2015, clinical data: 2011–2014. neurosurgical procedures. during a weekday after hours or weekends were more USA134 likely to experience complications (p= 0.0227); & had an increased mortality.
TABLE 21: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Neurology patients
Author, Patient number and Year, Study Design Main Findings Type Country Campbell, Prospective 45,726 stroke patients There was a stronger association with weekend admission & mortality 2014, UK 68 observational study of a admitted (23,779 which was associated with a 14% higher odds for 30-day mortality. large clinical dataset of admitted out of hours) acute stroke admissions: to 130 hospitals in 2010–2012. England. Crowley, Retrospective cohort 13,821 patients with Increased risk of death (30-days) for weekend admissions in patients with 2009a, USA. study of administrative diagnosis of ICH ICH (OR: 1.15; 95% CI: 1.05–1.25). 76 data: 2004. (26.8% weekend admissions). Fang, 2010, Retrospective cohort Consecutive patients Stroke fatality is higher with weekend compared to weekday admission, Canada. 46 study of administrative with acute stroke or TIA even after adjustment for case mix. The all cause 7-day fatality rate was data: 2003–2008. (20,657 patients). higher in patients seen on weekends compared to weekdays (8.1% v. 7.0%) even after adjustment for age, sex, stroke severity, & comorbid conditions (OR: 1.12; 95% CI: 1.00–1.25). Hasegawa, Prospective 1,134 consecutively Weekday admission was significantly associated (HR: 1.385 [1.087– 2005, Japan. observational study of admitting patients 1.764]) & case fatality (HR: 0.477 [0.285–0.798]). 48 clinical data: 2000– (multicentre – 10 2001. centres). Niewada, Prospective 19,667 ischemic Significant differential outcome (OR: 1.13). More patients admitted on Poland. 119 observational study of stroke (5,924 (30.1%) weekends died during hospitalisation or had poor outcome at discharge clinical data: 2004– weekend admissions) than weekday patients (15.9% & 59.8 v. 14.1% & 55.3%) respectively. 2005. from (73 centres). When corrected for clinical severity & effect persisted. Ogbu. 2011, Retrospective cohort 82,219 ischemic stroke Weekends represent a period of increased death risk for ischaemic stroke Netherlands. study of administrative admissions to 115 patients in the Netherlands. A higher 7-day death risk for weekend admission, 122 data: 2000–2004. hospitals. when compared to weekday admission was seen (OR: 1.27; 95% CI: 1.20– 1.34). However, this increased risk appears to represent an exacerbation of an underlying night-time risk present during the weekdays. Palmer, 2012, Retrospective cohort 93,621 (23,297 (24% Significant differential outcome (OR: 1.26) performance poorer at UK. 53 study of administrative weekend) stroke weekend on 5 of 6 metrics. The rate of 7-day in-hospital mortality for data: 2009–2010. admissions. Sunday admissions was 11% (adjusted OR: 1.26; 95% CI: 1.16–1.37 with Monday as a reference) compared with a mean of 8.9% weekday admissions. Reeves, Retrospective analysis of 187,669 acute ischemic Off-hour presentation was associated with increased risk of dying 2009, USA. 72 volunteer clinical stroke stroke (94 000 out in-hospital, although the absolute effect was small for ischemic stroke database: 2003–2007. of hours) & 34,845 (adjusted OR: 1.09; 95% CI: 1.03–1.14) & moderate for haemorrhagic acute haemorrhagic stroke admissions (adjusted OR: 1.19; 95% CI: 1.12–1.27). stroke (27,710 out of hours) admissions who presented to emergency departments form 857 hospitals that participated in a stroke program.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 25 TABLE 21: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Neurology patients
Author, Patient number and Year, Study Design Main Findings Type Country Saposnik, Retrospective data of 26,676 ischemic stroke Stroke patients admitted on weekends had a higher risk adjusted mortality 2007, USA. 125 prospectively collected patients (606 sites). than did patients admitted on weekdays. (OR: 1.14; 7-day mortality). administrative registry Weekend admissions data: 2003–2004. 6,629 (24.8% of all admissions). Sharp, 2013, Retrospective analysis 4,225,973-in-patient Patients more likely to die when admitted through the ED at the weekend USA. 159 of administrative data: admission via ED. US with 28 (adjusted) for 50 clinical groups. Individual characteristics used 2008. hospitals. for analysis did not explain the reasons for difference (DRG, patient insurance/ income status & or medical care or staffing patterns). Differential outcome across top 10 diagnosis (unadjusted OR: 1.073; 95% CI: 1.06–1.08; adjusted OR: 1.026; 95% CI: 1.005–1.048). Smith, 2010, Retrospective cohort 274,988 ischaemic Characteristics associated with in-hospital mortality were age, arrival Canada. 126 study of administrative stroke patients from mode, history of atrial fibrillation, previous stroke, previous myocardial data: 2001–2007. 1,036 hospitals. infarction, carotid stenosis, diabetes mellitus, peripheral vascular disease, hypertension, history of dyslipidaemia, current smoking, & weekend or night admission. Tung,2009, Retrospective cohort 34,347 ischemic stroke Weekend admissions were associated with increased 30-day mortality. Taiwan. 143 administrative data: patients (8,055 weekend Stroke patients admitted on weekends had a higher 30-day risk adjusted 2005. admissions) from 245 mortality than those admitted on weekdays. hospitals.
TABLE 22: Data extraction of studies reporting no impact from admission out of hours on outcomes – Neurology patients
Author, Patient number and Year, Study Design Main Findings Type Country Albright, Prospective collection 2,211 stroke patients There was no significant difference found for in-hospital mortality in any 2009, USA. - retrospective analysis admitted to 2 centres stroke category when comparing weekend to weekday admissions (OR: 19 of administrative data: (800 weekend 1.01; 95% CI: 0.74–1.39; p= 0.936). Patients admitted on weekends did 2004–2006. admissions). not have a higher rate of in-hospital mortality or 90-day mortality when compared to weekday admissions. Patients admitted on weekends also did not have worse functional outcome at discharge or at 90-days. Albright, Retrospective analysis 2,090 (720 weekend Weekend admission was not a significant independent predictor of 2012, USA. of administrative data: admission within 6 in-hospital mortality (8.4 v. 9.9% p= 0.056), or 90-day mortality (18.2 v. 20 2002–2009. hours of stroke onset 19.8%; p= 0.680). No significant difference in treatment rates between to CSC (8 sites). weekday & weekend groups. Almekhlafi, Retrospective cohort 110 patients (56 out of No significant difference in-hospital mortality (2% for those treated on 2014, study of clinical data: hours) (single-site). weekends & evenings v. 5.6% for those treated during working hours (p= Canada. 129 2011–2012. 0.3). Independent functional recover at 90-days was noted in 64.3% of those treated on evenings or weekends & 52.1% of those treated during working hours (p= 0.2). Crowley, Retrospective cohort 5,667 patients Unlike the groups other study, weekend admission was not a statistically 2009, USA. study of administrative admitted with SAH significant independent predictor of death in the study population at 7-days 75 data: 2004. on weekends. US (OR: 1.07; 95% CI: 0.91–1.25), 14 days (OR: 1.01; 95% CI: 0.87–1.17), hospitals. or 30-days (OR: 1.03; 95% CI: 0.89–1.19). Weekend admission is not associated with significantly increased short term mortality risk (30-days) among patients admitted with subarachnoid haemorrhage (OR: 1.03; 95% CI: 0.89–1.19). Jiang, 2011, Retrospective cohort 313 patients with ICH Found that weekend admission was not a statistically significant predictive China. 114 analysis of clinical (single-site). factor of in-hospital mortality (p= 3.15) or functional outcome (p= 0.128). data: 2007–2009. Sato, 2015, Post-hoc analysis of an 2,794 (1,770 (63%) off- Off-hour admission was not associated with risk of poor outcome at 90- Australia. 36 international multicentre hours) ICH admissions days (53% off-hour v. 55% on-hour p= 0.49). There were no significant open, blinded endpoint to stroke centre. associations after adjustment for various baseline risk factors & clinical RCT: 2008–2012. features (OR 0.92; CI: 0.76–1.12) for off-hour admission compared to the on-hour admission patients. Similar no clear associations were observed for death (OR 1.29; CI: 0.94–1.77) or disability (OR 0.86; CI: 0.72–1.03). Zhang, Retrospective study of 183 SAH. (single-cite). Logistic regression model, weekend admission was not an independent 2011, China. clinical data: 2006– predictor of higher in-hospital mortality (OR: 177; 95% CI: 0.83–3.77) after 156 2009. SAH.
26 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project TABLE 23: Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Neurology patients
Author, Year, Study Design Patient number and Type Main Findings Country Bejot, 2013, Retrospective cohort 5,864 (1,465 out of hours) 30-day mortality was greater for patients with stroke of France. 81 study administrative cases of TIA. Dijon, France TIA onset during weekends/ holidays than for those with data: 1985–2010. hospitals. events occurring during weekdays (15.2% v. 12.1%). Excess mortality in patients with stroke of TIA onset during weekends/ holidays was observed for the period 1985–2003 (1,821 v. 14%) but was no longer found for the period 2004–2001 (8.4% v. 8.3%). Bray, 2014, Prospective cohort 56,666 patients in 103 stroke In multivariable analysis, patients admitted on a weekend UK. 55 study of clinical data: units, NHS England hospitals. to stroke unit with 1.5 nurses / 10 beds had an estimated 2011–2012. adjusted 30-day mortality risk of 15.2% (OR: 1.18; 95% CI: 0.77–0.93). Hoh. 2010, Retrospective cohort 599,087 emergency Patients admitted out of hours compared with regular hours USA. 90 study - administrative admissions for ischemic stroke were more like to receive thrombolytic (OR: 1.114) & have data: 2002-2007. - 159,906 weekend admissions slightly longer LoS (OR: 1.021). No difference for in-hospital & 439,181 weekday mortality or discharge disposition. admissions. US hospitals. Inoue, 2015, Retrospective cohort 47,885 stroke patients. There was no difference in mortality between week day & Japan. 58 study of administrative weekend for patients that were admitted to the SICU on data: 2010–2011. a weekend (10% v. 9.9%). The in-hospital mortality was significantly higher among patients admitted on a weekend to the GMW (7.9 % v. 7%). Karlinski, 2013, Retrospective cohort 1,330 intravenous thrombolysis Significant differential outcome (OR: 1.07). Increases in Poland. 135 study – administrative patients (448 weekends; 868 mortality associated with out of office hours admissions for (stroke registry) data: out of office hours & 105 night stroke were most pronounced in smaller hospitals. 2003–2006. admissions) 27 Polish stroke centres. Kazley, 2010, Retrospective cohort 78,657 acute ischemic stroke No impact of weekend admission on outcome (OR: 1.024 USA. 60 study administrative & 20,101 haemorrhagic stroke case mix adjusted mortality). data: 1998–2006. admissions from all hospitals in Virginia. McKinney, Retrospective analysis 134,441 patients admitted with In-hospital & 30-day mortality rates were also increased 2011, USA. 71 study of administrative primary diagnoses of cerebral for patients admitted on weekends Mortality 90-days after data: 1996–2007. infarction (27.8% (37,321) admission was significantly higher for patient admitted on admitted on weekends) New weekends than on weekdays (17.2% v. 16.5%; p= 0.001). Jersey hospitals. O’Brien, 2011, Observational study of 929 stroke patients in 4 US Overall risk of 28-day mortality was 9.6 % for weekday USA. 121 clinical data: 2004. communities. strokes & 10.1% for weekend strokes. However, in a model controlling for patient demographics, risk factors & event year, weekend arrival was not associated with 28-day mortality (OR: 0.87; 95% CI: 0.51–1.50). Turin, 2008, Retrospective cohort 1,578 registered first ever For all strokes, the 7-day case fatality rate based on the Japan. 144 analysis study of cerebral infarction & cerebral hospital admission day was 9.5% (95% CI: 6.8-13.1) for administrative data: haemorrhage stroke cases. weekend admissions & 7.3% (95% CI: 6.0-8.9) for weekday 1998–2003. Hospitals in central Japan. admissions. However, case fatality rates based on the onset day were 7.2% (95% CI: 5.1-10.0) for weekend onset & 8.0% (95% CI: 6.6-9.8) for weekday onset. The 28-day case fatality rate for the weekend admission group was 14.7% (95% CI: 11.3-18.8) and for the weekday admission group it was 10.1% (95% CI: 8.5-11.9). In contrast, the 28-day case fatality rate for the weekend onset group was 11.3% (95% CI: 8.6-14.7) & for the weekday onset group it was 11.0% (95% CI: 9.3-13.0). This phenomenon was observed mainly for cerebral infarction & to some extent for cerebral haemorrhage.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 27 Paediatrics and Obstetrics We identified five obstetric and paediatric studies from the database searches and snowballing. All of these studies were analysis of administrative datasets except for one case series analysis. Each of the studies investigating administrative datasets found a statistically significantly greater risk of mortality for patients admitted out of hours.The case series study concluded no impact of out of hours admissions on outcome. The characteristics and findings of the included reviews and primary studies are summarised in Tables 23–24.
All five papers reviewed in-hospital mortality only. No papers reviewed LoS or compared delay in treatment to patients admitted within traditional working hours. One 149 study was a single-site study with the remaining four 18 99 140 146 analyses from mult-site centres.
TABLE 24: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Paediatrics and Obstetrics patients
Author, Year, Study Design Patient number and Type Main Findings Country Auger, 2015, Retrospective analysis of 55,383 hospitalisation at Children admitted at the weekend had significantly higher USA. 149 administrative data: a tertiary care children’s odds of unplanned readmission compared to children 2006–2012. hospital. admitted on weekdays (adjusted OR: 1.09; 95% CI: 1.004–1.18). In contrast, being discharged at the weekend was not associated with readmission. In conclusion, children admitted at the weekend have higher rates of 30-day unplanned readmission than children admitted during the week, suggesting care differences at the weekend related to initial clinical management rather than discharge planning. Goldstein, Retrospective cohort 439,457 paediatric patients After multivariate adjustment & regression, patients 2014 USA. 140 analysis of administrative who underwent a range of undergoing a weekend procedure were more likely to die data: 1988–2010. surgical procedures. (OR: 1.63; 95% CI: 1.21–2.20). Hamilton, Retrospective descriptive 111,749 Births to Teenage Significant weekend mortality confined to African Americans 2003, USA. 146 design study of mothers (397 neonatal & Hispanics (not Caucasians) (OR: 1.42). administrative data: deaths). 1999–2000. Pasupathy, Retrospective descriptive Live born term singletons Delivering an infant outside of the normal working week 2010, design study of with cephalic presentation. was associated with an increased risk of neonatal death Scotland. 18 administrative data: Perinatal deaths from at term ascribed to intra-partum anoxia. Risk of neonatal 1985–2004. congenital anomalies death was 4.2 per 10,000 during the normal working week excluded. Final sample & 5.6/10,000 at all other times (out of hours) (unadjusted comprised 1,039,560 live OR: 1.3; 95% CI: 1.1–1.6). births.
TABLE 25: Data extraction of studies reporting no impact from admission out of hours on outcomes – Paediatrics and Obstetrics patients
Author, Year, Study Design Patient number and Type Main Findings Country Gould,2003, Case Series of live births 16,150,41 live births. Observed neonatal mortality increased was 2.80/1000 for USA. 99 (weigh >500g) 1995–1997. Hospitals, California, US. weekday births & 3.12/1000 for weekend births (OR: 1.12; 95% CI: 1.05-1.19; p= 0.001). But after adjusting for birth weight, the increased odds of death for infants born at the weekends were no longer significant.
28 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project Orthopaedics We identified six papers that were included in the review. Two papers each reported a difference in mortality for patients admitted out of hours, no difference in outcome or mixed outcomes. Three papers interrogated administrative datasets with the remaining three analysing clinical datasets. One study was a prospective population cohort study. The characteristics and findings of the included reviews and primary studies are summarised in Tables 25–27.
An additional eight papers 4 6 7 9-11 24 107 123 are not included in the below summary table, as they examined in detail above within the above in Tables 3–5.
Four studies 57 61 150 154 investigated in-hospital mortality. Patient outcomes at 30-days were also investigated in four studies. 51 61 77 154 The variable, delay in treatment was tested in one study,61 finding no difference in treatment delay for patients admitted outside of hours compared to other times. Both of these studies were an interrogation of clinical data. No studies investigated the difference in LoS. A breakdown of these variables are summarised in Appendix 5.
Hip fractures related to fragility account for a significant and clinical and economic burden on our health system, especially in an aging population. The studies reviewed indicated that patients admitted with hip fracture often have multiple co-morbidities and can present with concomitant medical pathologies such as ischaemic heart disease, electrolyte imbalances, renal impairment and sepsis. The studies highlight that the effective management of these, medical optimisation and access to timely surgery are key factors in the effective treatment of hip fractures and prevention of further complications.
TABLE 26: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Orthopaedics patients
Author, Year, Study Design Patient number and Type Main Findings Country Foss, 2006, Prospective 600 (118 weekend & 150 Found significantly higher rate of mortality for those admitted over Netherlands. 77 cohort study of holiday periods) consecutive holiday periods. Both 5-day & 30-day postoperative mortality were clinical data: patients with patients presenting significantly higher in patients admitted during holiday periods than 2002–2004. with acute hip fracture. during weekends & weekdays, 8.0% v. 2.5% & 1.8%, respectively (p= 0.01) & 19.3% v. 12.7% & 11.1%, respectively (p= 0.05). In a multivariate analysis, admission during holiday periods was still a significant independent risk factor for both 5-day (4.34; 95% CI: 1.74–10.8) & 30-day mortality (OR: 1.84; 95% CI: 1.08–3.12). Thomas, 2014, Retrospective 2,989 fractured neck of femur Researchers found that (whether managed surgically or UK. 154 cohort study of (single-site). conservatively), patients were more likely to die as an inpatient clinical data: when admitted at the weekend (OR: 1.4; 95% CI: 1.02–1.80; p= 2009–2013. 0.032).
TABLE 27: Data extraction of studies reporting no impact from admission out of hours on outcomes – Orthopaedics patients
Author, year, Study design Patient Number and Type Main Findings country Daugaard, Retrospective 38,020 fractured neck of femur. The mortality rate for patients admitted during weekends or public 2012, cohort analysis of Entire Danish population. holidays or at night was similar to those four those admitted during Netherlands. 57 administrative working days. data: 2003–2010. Mathews, Retrospective 796 patients admitted to single- Weekend admission was not associated with increased time-to- 2016, UK. 61 cohort analysis site hospital (235 weekend surgery or mortality in hip fracture patients. Demographic factors of prospectively admissions) with hip fracture affect mortality. Weekend admission was not associated with an collected clinical (edge of femoral head & 5cm increased time-to-surgery (p= 0.975), 30-day mortality (p= 0.842) or data: 2009–2011. below the lesser trochanter). 120-day mortality (p= 0.425).
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 29 TABLE 28: Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Orthopaedics patients
Author, Year, Study Design Patient number and Type Main Findings Country Boylan, 2015, Retrospective analysis 344,989 fractured neck of femur. Compared with patients admitted on weekdays. Patients USA. 150 of administrative admitted on weekends had lower mortality (OR: 0.94; 95% data: 1998–2010. CI: 0.89–0.99) & shorter mean hospital stay (estimate 3.74%; 95% CI: 3.4–4.08); but did not differ in risk of perioperative complications (OR: 1.00; 95% CI: 0.98–1.02). The study data did not support a differential outcome among hip fracture admissions in the US. Kristiansen, Cohort study, 25,305 Danish patients undergoing When comparing admission on weekdays (evenings & 2016, administrative data: hip fracture surgery, aged >65 nights v. days), off-hours admission was associated with Netherlands. 51 2010–2013. years. a lower risk of surgical delay (adjusted OR: 0.75; (95% CI: 0.66–0.85) while no differences in 30-day mortality was found (adjusted OR: 0.91; 95% CI: 0.80–1.04). When comparing admission during weekends with admission during weekdays, off-hours admission was associated with a higher risk of surgical delay (adjusted OR: 1.19; 95% CI: 1.05–1.37) & a higher 30-day mortality risk (adjusted OR: 1.13; 95% CI: 1.04–1.23). The risk of surgical delay appeared not to explain the excess 30-day mortality.
Trauma We identified ten papers investigating the difference in outcomes for patients admitted out of hours compared to other times admitted with a trauma. Two papers concluding an increased risk of mortality, seven found no difference and one concluded mixed outcomes. Eight studies analysed clinical data with two interrogated administrative data. The characteristics and findings of the included reviews and primary studies are summarised in Tables 28–30.
All ten studies were retrospective cohort studies all investigating in-hospital mortality. Three 56 70 94 were multi-centre studies all interrogating clinical data. Five 64 70 86 118 123 of the papers analysed the out of hours time frame as nights and weekends together with four 56 94 132 133 analysing the outcomes of patients admitted at nights and weekends separately. Only two papers 70 133 reviewed patient severity of illness, neither finding how unwell the patient was on presentation impacted on the risk of mortality. Three studies 56 86 94 investigated whether LoS was greater for patients admitted out of hours, two 56 94 finding no difference. A breakdown of these variables are summarised in Appendix 5.
TABLE 29: Data extraction of studies reporting an impact from admission out of hours with higher mortality or other adverse outcomes – Trauma patients
Author, year, Study design Patient Number and Type Main Findings country Mitra, 2014, Retrospective analysis of 398 patients with ATC (197 The out of hours model of care was associated with Australia. 118 clinical data; 2006–2011. (49.5%) presented after worse outcomes among some of the most critically ill hours). trauma patients. Mortality among patients presenting after hours was 43.1%, significantly higher than among those presenting in hours (33.1%; p= 0.04). Following adjustment, after-hours presentation was significantly associated with higher mortality at hospital discharge (adjusted OR: 1.77; 95% CI: 1.1–2.87). Schneider, Retrospective cohort 38,675 individuals aged Weekend patients demonstrated 14% increased risk of 2012, USA. 86 analysis of (NIS database) 65–89 years with head mortality (OR: 1.14; 95% CI: 1.05–1.23). Older adults with administrative data: trauma. substantial head trauma admitted on weekends are less 2006–2008. severely injured, carry less comorbidity, & generate similar total charges compared with those admitted on weekdays. However, after accounting for known risk confounders, weekend patients demonstrated 14% greater odds of mortality. Mechanisms behind this disparity must be determined & eliminated.
30 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project TABLE 30: Data extraction of studies reporting no impact from admission out of hours’ on outcomes – Trauma patients
Author, Year, Study Design Patient number and Type Main Findings Country Arbari, 2005, Retrospective cohort study of 30,686 admissions to a After controlling for confounding variables, no difference USA. 148 clinical data: 1994–2002. Level 1 Trauma Centre. in mortality or LoS was observed in association with weekend or night admission. Buse, 2004, Retrospective cohort study of 1,044 trauma patients Time of presentation (night time v. day time weekend v. USA. 132 clinical data 1995–2001. admitted to single-site trauma weekday, month of year, & year) was not associated with centre. in-hospital mortality after adjustment of illness & other confounding variables. Carmody, Retrospective cohort study of 8,015 trauma admissions to Subset analysis were performed for time of admission, 2002, USA. 133 clinical data: 1994–1996. a large teaching centre LA, day or admission & mechanism of injury: morning v. California. night admission; weekday v. weekend; least busy-day (Tuesday) v. busiest day (Sunday); week-night v. weekend night admission. Penetrating trauma: morning v. night; blunt trauma: morning v. night. None of the 6 comparisons showed a significant difference in mortality. Carr, Retrospective cohort 90,461 trauma patients 32 No difference. Patients presenting on weekends were less 2011,USA. 56 analysis of clinical data: trauma centres, Pennsylvania, likely to die presenting on weekdays (OR: 0.89; 95% CI: 2004–2008. US. 0.81–0.97). Laupland, Retrospective cohort study of 4,000 major trauma patients Neither admission of weekends nor on evenings or nights 2009, clinical data: 2002–2006. Calgary. increased the risk for in-hospital mortality. The time of Canada. 70 admission during the day or day of the week does not influence the risk for adverse outcome & may reflect highly developed multi hospital acute care & trauma system. Taira, 2009, Retrospective cohort study 25,572 patients (17,625 No difference in ICU LoS, hospital LoS, or mortality (p= USA. 94 – (secondary analysis) of arriving off-hours) 700 trauma 0.546) for those admitted during off-hours compared with clinical data: 2002–2006. facilities included with injury weekday admits. After adjustment hospital type, off-hours mechanism of burn. admission was not predictive of mortality (OR: 1.06; 95% CI: 0.91–1.23). Nandyala, Retrospective cohort study 34,122 patients who Found the mortality rate was not significantly different 2012, USA. 64 of administrative data: undergone cervical fusion among the weekend patients. There were no significant 2002–2011. for cervical spine trauma. US differences in mortality based on the admission day for hospitals. any surgical approach.
TABLE 31: Data extraction of studies reporting mixed outcomes for patients admitted out of hours – Trauma patients
Author, Year, Study Design Patient number and Type Main Findings Country Ono, 2015, Retrospective cohort study of 805 patients (426 Off-hours presentation was associated with longer ER Japan. 123 clinical data: 2002–2013. presentations out of hours). stays for patients with systolic blood pressure <90mmHg on admission (p= 0.021), ISS >15 (p= 0.047), & pelvic fracture requiring trans arterial embolization (p<0.001). Off-hours presentation was also associated with increased risk of adverse events in the ED (OR: 1.7; 95% CI: 1.1 -2.7; p= 0.020). After adjustment for confounders, an increased risk of adverse events (OR: 1.6; 95% CI: 1.1–2.7; p= 0.049) persisted, but no differences were detected in mortality (p= 0.80) & unexpected death (p= 0.44) between off-hours & business hours.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 31 Chapter 6: What influences the inequitable outcome for patients admitted out of hours and at the weekend (when there is one) is undecided
A number of potential causes of a differential outcome for patients admitted out of hours are proposed by the authors of papers examining this issue These suggestions are made broadly speaking, based on the expert opinion in one of two scenarios:
1. That the presence of certain human resource, system or process factors mitigated against there being a difference in outcome; or
2. That the absence of those features is postulated to be associated with differential outcomes out of hours.
From review of the literature, when an effect was present, causation was postulated but not proven. Furthermore, no studies were found that specifically evaluated an intervention to modify or reduce a differential outcome should one have been previously identified. As such any hypothesised factors, that may explain differences in outcome, have not as yet been formally evaluated.
Patient differences, out of hours access to diagnostic investigations and treatments, and staffing profiles have all been implicated in either the association with or mitigation of a differential outcome. The largely retrospective nature of all studies, clinical and administrative, suggests to the reviewers that these are hypothesis generating for factors that may impact upon outcomes out of hours rather than being definitive in and of themselves.
This review has identified that the majority of studies that corrected for severity based on clinical data did not identify an association between out of hours admission as opposed to those studies that only corrected for comorbidities.
It is impossible to make any conclusion regarding the role that delay to treatment and investigations plays in increasing mortality of those admitted out of hours due to the paucity of research in this area.
Finally, staffing appears to be an important factor in contributing to the differential outcome. However, determining exactly which part of staffing (medical, nursing, allied health) remains elusive with most of the studies based on specific models of care (see below) making it difficult to determine which components contribute to the effect more than others. Patient acuity and/or severity of illness An increased patient acuity and/or severity of illness has been proposed as an influence in the inequitable outcome for patients admitted out of hours.
To determine the impact of illness acuity and/or severity (if any), studies have been separated into those that used adjusted clinical data and those that used administrative data are presented. The adjustments included physiological variables and other severity scores such as APACHE, stroke severity, and trauma severity scores. Studies using administrative data do not have the same access to these admission variables and use comorbidities or socioeconomic measures to adjust for severity.
The majority of studies using clinical data to adjust for illness severity did not identify an association between out of hours admission and mortality. The studies that did identify an association were mostly based on administrative data and corrected only for comorbidities.
Of the 146 studies, we have examined in more detail 20 administrative dataset studies and 36 clinical dataset studies proposed that patient acuity and/ or severity of illness may have been an influence on the outcomes.The characteristics and findings of the included reviews and primary studies are summarised in Tables 31 and 32.
Five papers from the 11 administrative dataset studies investigating the outcomes for admitted cardio respiratory patients indicated that diagnosis and severity of diseases and conditions may be a contributor of differences of admission, LoS and mortality. Of the 15 clinical dataset papers concerning cardio respiratory patients, seven considered severity of illness. One study 83 found higher complication rates admitted off-hours. Other studies of AMI did not demonstrate higher complication rates measured by major adverse cardiovascular events (re-infarction stroke repeat target vessel revascularisation of severe bleeding). 22 69
32 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project Of the 30 papers investigating outcomes of patients acutely admitted for neurosurgery, stroke and other neurological diagnoses four administrative data set studies and seven clinical dataset studies indicate severity of illness as a possible factor.
From a total of 15 studies of gastroenterology patients, four 43 49 74 49 of eight administrative dataset studies and four 27 35 42 91 of six clinical data set studies determined patient acuity and or severity as a factor. Of the four clinical dataset studies, three 27 35 91 found no difference in outcome for patients once adjusted for these factors. For example, one 91 comparing admissions of patients on weekends compared to other days of the week had higher prevalence of coagulopathy and were likely to be admitted emergently.
Fourteen studies of patients admitted to the ICU from the 20 papers reviewed from the database search indicated severity of illness and/or patient acuity to be a factor contributing to the outcome. One study 103 found that patients admitted during off-hours were less sick. This contrasts with three 26 78 108 papers reviewed finding that patients admitted out of hours were not the same as those admitted during regular ‘day’ shifts, finding that patients admitted out of hours were older, more critically ill and frequently required supportive procedures.
One study 132 indicated that emergency surgery, with possibly more dangerous outcomes could more commonly be carried out on a weekend than a weekday.
Any effect might partly be explained by the average admitted patient being sicker at the weekend with patients admitted out of hours had predominantly more complex disorders and as such had higher associated mortality. 4 46 91 125 Staffing number and seniority It is reasonable to hypothesise that differences in outcomes for patients admitted out of hours may be due to a lower quality of clinical services at these times.
There is variable evidence that working patterns are a factor in the differences in outcome for patients admitted out of hours. Very few papers were identified to compare different physician staffing patterns or test whether one staffing model was superior to another in terms of the effect of admission or time of day on patient outcomes. Further, from the papers reviewed there was not enough information about the severity of a patient diagnosis and the time staff resources were needed in the care of that patient. The studies referred to acuity and quality of care as the workload and burden of care needed for severity of illness. In theory, quality of care with regard to the acuity of severity of illness with each patient diagnosis would produce a different workload burden for staff. As such more work is needed on acuity processes and patient outcomes in order to determine how patient diagnosis affects the workload of staff.
Of the 146 studies, we have examined in more detail 33 papers that identified that staffing may be a contributory factor, however, very few papers investigated these components individually even though they are frequently mentioned as possible influences. The majority of studies investigating staffing and seniority factors looked more at whole teams or units providing a specific model of care for a particular condition. Consequently, specific models of care are reviewed in more detail in a separate section. The characteristics and findings of the included reviews and primary studies are summarised in Tables 33 (a–f) and 34 (a–e).
A number of papers identified that staffing may be a contributory factor to differences in outcome for patients admitted out of hours. Many postulate that the differences in care may be associated with nursing ratios and senior consultant direct patient care.
One 81 paper identified mixed outcomes and one paper 79 purporting that nurse to patient ratios may be a factor however no difference in outcomes for admitted patients was found. One study 68 investigated inadequate staffing numbers or seniority. The research broke down stroke care into a number of bundles of which nursing ratios and senior consultant presence was analysed. Interestingly, they reported that 24/7 consultant ward rounds had no effect on stroke mortality with out of hours admission but the nursing ratio did. They identified that a lower ratio of nurses to patients had a higher mortality. Another study 106 reported that competing tasks and the less experience of the treating physicians in the rural setting, who may be general practitioners as opposed to stroke specialist, may be the cause of increased mortality of afterhours admissions. Treatment delay It is plausible to think that delay in appropriate diagnosis and management in several medical conditions may ultimately affect the potential success of any intervention or specified medical treatment.
There is a paucity of studies directly investigating this component. Of the 146 studies, we have examined 22 papers in more detail indicating that treatment delay was or was not a factor to impact on the outcomes of patients admitted out of hours. The characteristics and findings of the included reviews and primary studies are summarised in Tables 35 (a–g) and 36 (a–d).
The majority of these papers analysed clinical datasets. Of these studies, five 42 68 82 119 determined a delay in treatment for patients admitted out of hours and concluded that this delay in treatment impacted on the outcome. six 27 36 61 91 128 129 determined treatment delay for patients admitted out of hours, yet however this did not appear to impact on outcome.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 33 34 TABLE 32 (a–e): Data extraction of administrative studies identifying patient acuity as a contributory factor of predicting unequal health outcomes for patients admitted out of hours Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project a. Cardio Respiratory patients Difference Author, Year Comment in outcome Country Yes Aujesky, Several explanations are possible for the association between weekend admission & increased 30-day mortality, particularly among the most severely ill patients. 2009, USA. 73 The higher mortality for patients hospitalised on weekends was driven by the increased mortality rate among patients within the highest severity of illness risk class at presentation. Mixed Becker, 2008, The study found that patients are no more likely to have been hospitalised in the year prior, or Charlson Indices that are statistically indistinguishable from those of USA. 22 weekday patients. Researchers also found a slight spike in the number of AMI patients admitted on Mondays, the absence of a weekends effect suggests that this is unlikely to be the result of less severely ill patients experiencing weekend symptom onset deferring admission until Monday. Yes Deshmukh, Patients with 3 or more co-morbidities had the greatest association with in-hospital mortality compared to all other parameter. 2012, USA. 80 Yes Kostis, 2007, In the investigation of 4 cohorts with over 230,000 patients during 1987–2002, for possible differences in outcome for patients admitted out of hours . They reported USA. 50 4.8% higher mortality rates at 1-month in patients admitted on weekends v. week-days, which persisted even after adjustment of demographic variables. The magnitude of this increase in mortality how-ever, was reduced to 2.3% and became non-significant (p= 0.09) once analysis accounted for invasive cardiac procedure. Yes Kruth, 2008, The study found that in a multivariate analysis after adjusting for age, gender, previous MI, diabetes, hypertension, pre-hospital delay and tachycardia, we identified a Germany. 69 significant increase of in-hospital mortality on weekends. Also, there was a non-significant increase of in-hospital mortality at nights as compared with ‘‘on’’-hours.
b. Gastroenterology patients Difference Author, Year Comment in outcome Country No Jairaith*, The study found that patients presenting at weekends were more likely to present with shock and hematemesis and to receive blood transfusion compared with 2011, UK. 49 weekday presentations; and that there was no difference in the risk adjusted mortality of patients presenting with UGIB at weekends compared with weekday. Yes Button*, 2011, The study found that people admitted on weekends and public holidays were slightly but significantly younger (63.4 .v 64.4 years; p= 0.020), were discharged slightly UK. 74 sooner (overall median LoS = 4 days for both; but p= 0.022) and had similar levels of co-morbidities recorded compared with patients who were admitted during the week. Yes Dorn,* 2010, The study found that patients admitted with UGIB over the weekend are likely sicker. Based on a Bonferroni correction for multiple testing (a=0.05/29), those admitted USA. 43 over the weekend were significantly more likely to have baseline coagulopathy, fluid and electrolyte disorders, and a history of alcohol abuse, and less likely to have chronic blood loss anaemia. However, despite statistical significance on these effects of out of hours admission was stronger (HR: 1.17: 95% CI: 1.03–1.32). c. Haematology-Oncology and other medical specialities patients Difference Author, Comment in outcome Year Country No Laupland, While we did not adjust for severity of disease in our study, our observation that neither crude nor adjusted risk was increased argues against a significant after hours 2010, effect. The present study was not powered to exclude very small but potentially statistically significant differences in risk. However, given that no excess risk was Canada. 138 observed in this study that included all first admissions with community onset bloodstream infection in a population of more than a million people over a 9-year period, argues against a clinically significant effect. d. Medical patients Difference Author, year Comment in outcome country Yes Aylin, 2010, UK. 6 The study concluded that there could have been differences in case mix between patients admitted during the week and weekends, but acknowledge that all confounding factors were accounted for. d. Medical patients Difference Author, year Comment in outcome country Yes Aylin, 2013, UK. 7 The study investigated ‘risk of procedure’, but as this was not based on clinical severity, it might not have captured the complete risk profile of patients. The researchers purport however that given that our analysis suggests that patients operated on at the weekend were likely to undergo lower risk procedures, however, again this seems unlikely to account for the findings. Mixed Clarke, 2010, Researchers purported that it can also be speculated that we found a significant differential outcome on myocardial infarction compared with COPD because Australia. 24 myocardial infarction has a relatively higher baseline mortality and thus severity. However, the researchers also acknowledge that they cannot exclude the possibility that patients admitted on weekends were sicker than those admitted on weekdays. The only potential markers of disease severity available in state wide Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project database were ventilation received (relevant for COPD and intracerebral haemorrhage) and presence of acute respiratory failure (relevant for COPD). These did not vary much across weekend v. weekday admissions and were adjusted for in the logistic regression. Therefore, we believe there was little, if any, confounding. Also, morbidities we were unable to distinguish between pre-existing conditions and complications because this information was not available on the entire data set. Mixed Maggs, 2010, UK. The study found that patients admitted at the weekend who survived the first seven days do not have a higher subsequent mortality rate, but those admitted on 147 Monday take days, at night or out of hours do appear to have a worse outcome even if they survive the first seven days. We postulate that these differences in late mortality are more likely to be explained by variation in initial illness severity, since any in-hospital factors around the time of admission would be expected to have less impact after seven days. Yes Freemantle, 2012, The study found that the cohort of patients admitted on weekend were patients who would otherwise had they been less ill, have had their admission postponed UK. 10 until a weekday. Yes Freemantle, 2015, Although fewer hospital admissions occur at the weekend, patients admitted on Saturday and Sunday are sicker and face an increased likelihood of death UK. 11 within 30-days even when severity of illness is taken into account. e. Neuroscience patients Difference Author, Year Comment in outcome Country No Crowley, 2009, The study did not have any reason to believe that the severity of patients presenting with SAH is different on weekends as opposed to weekday. USA. 75 No Bejot, 2013, The study found no differences between the two groups for either distribution of cerebrovascular event subtype or stroke severity (measured either by the France. 81 means of clinical proxies of using the NIHSS score) which is known to be the strongest predictor of a poor outcome in stroke patients. Yes Crowley, 2009, This study was unable to measure and adjust for potentially important confounders such as differences in cause of death, admission grade and stroke severity, USA. 76 level of consciousness, imaging findings, size of haemorrhage, the need for cerebrospinal fluid diversion, whether the ICH was evacuated, and graded outcome scale. Yes Palmer, 2012, UK. 53 The researchers we cannot evaluate how much, if any, of the differential outcome can be attributed to differences in stroke severity.
TABLE 33 (a–f): Data extraction of clinical studies identifying patient acuity as a contributory factor of predicting unequal health outcomes for patients admitted out of hours
a. Cardio Respiratory patients Difference Author, Year Comment in outcome Country No Al-lawati, 2012, Similar number of patients with diagnosis of ST-elevation (39 v. 41%; p= 0.14) and non-ST elevation (32 v. 33%; p= 0.57) MI were admitted on weekdays and UAE. 128 weekends, respectively. There were slightly more admissions with unstable angina on weekdays than weekends (29 v. 26%; p= 0.02). There were no significant differences between patients admitted on weekdays and those admitted on weekends in terms of age, gender preponderance (72% men), smoking, co-existing morbidities, prior history, anatomical site of MI, medication administered in the first 24 hours, rate of in-hospital complications and LoS in-hospital. 35 36 a. Cardio Respiratory patients Difference Author, Year Comment Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project in outcome Country No Fonarow, 2008, Patient characteristics were also similar when patients admitted on weekdays were compared with those admitted on weekdays. Patients admitted on USA. 145 weekends for HF did not appear to be sicker than those admitted on weekdays, as judged by the severity of symptoms and multiple prognostic variables. No Jneid,2008, USA. Patients arriving during off-hours were younger, were more likely to belong to minority groups, and had slightly higher body mass index. They also were less 59 likely to have a history of atrial fibrillation but more likely to have diabetes mellitus, heart failure, previous MI, and adult history of smoking and to present with STEMI. Yes Lairez, 2009, The study found no significant differences in the occurrence of recurrent MI, CABG, new PCI, stroke, or renal and vascular events between patients treated France. 52 with emergency PCI at night and during the day (2.2% v. 2.0%; p= 0.73) or between patients treated during off-hours and regular hours (2.2% v. 1.9%; p= 0.68 respectively). No Sorita,2014 USA. 83 This study found that patients with AMI admitted during off-hours were more likely to have STEMI, present with cardiogenic shock, and develop shock after presentation compared with those admitted during regular hours. No Magid, 2008, However, patients in the fibrinolytic and PCI cohorts who came to the hospital during off-hours were younger, more likely to be smokers, and less likely to have USA. 34 had a pre-hospital ECG than patients who came to the hospital during regular hours. Compared with patients who presented during regular hours, patients in the fibrinolytic cohort who presented during off-hours were more likely to be admitted to hospitals with cardiac surgery capability but less likely to be admitted to hospitals that used fibrinolysis for more than 90% of reperfusion cases. Finally, patients in the PCI cohort presenting during off-hours were more likely to be admitted to high-volume PCI centres and hospitals that used PCI for more than 90% of reperfusion cases than patients presenting during regular hours. No Sato, 2015, Patients with off-hour admissions were less likely to have had prior antihypertensive, antithrombotic and lip lowering treatment and also had lower GCS scores. Australia. 36 Researchers purported that this may be partly due to the difficulty in obtaining an accurate history off-hour, especially in patients with lower consciousness levels. Even though there could have been external factors causing these differences in baseline characteristics, it seems to suggest that patients who were critically ill were more likely to be admitted to stroke centres during off-hours. The researchers also found it intriguing that off-hour admissions were related to greater risk of death at 90-days in patients receiving surgical intervention within 7-days. Though off-hour admission might affect decision making over the indication of surgery and / or postoperative care, the strength of evidence is limited by the small number of clinical events. b. Gastroenterology patients Difference Author, Year Comment in outcome Country Yes De Groot, 2012, Patients admitted during the weekend more often presented with haematemesis and collapse and more often had a lower systolic and diastolic blood pressure Netherlands. 42 and tachycardia at presentation. Rates for re-bleeding and need for angiographic and surgical interventions were also higher in patients admit-ted during the weekend.
No Haas*, 2010, The study found that no mortality difference and no difference between groups – severity or presenting symptoms. USA. 27 No Myers*, 2009, The only clinically significant difference was a higher frequency of emergent admissions for patients hospitalised at the weekend v. patients hospitalised on a Canada. 91 weekday (68.2% v. 64.2%, respectively; p<0.0001). No Worni, 2012, No significant difference was detected in severity of appendicitis between weekday and weekend admission. USA. 35 c. Intensive Care patients Difference Author, Year Comment in outcome Country No Arabi, 2006, Elective admissions were excluded because they occurred mainly during weekdays and are typically of lower severity of illness. Admissions to the intermediate Saudi Arabia. 21 care unit, coronary care unit, and cardiac surgical ICU were also not included in the study. The study also shows that two-thirds of emergency admissions are actually admitted during the after-hours with no measurable differences in severity of illness among the three time periods. c. Intensive Care patients Difference Author, Year Comment in outcome Country No Ensminger*, Patients admitted to ICU at the weekend were sicker, as measured by the APACHE III score and predicted mortality rate. Patients admitted to the ICU at the 2004, USA. 26 weekend were less likely to be admitted from recovery rooms or operating rooms or to have a postoperative diagnosis. They were also more likely to require high risk monitoring or active treatment. The 3 most common ICU admission diagnosis groups for both on & off-hours admissions were cardiovascular, respiratory, and GI. The current study showed that the intensity of treatment and the ICU admission source, as well as the APACHE III predicted mortality rate, were independently associated with increased hospital mortality. (Subgroup analysis demonstrated increased severity on weekends but no effect on mortality after adjustment for severity)
Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project Yes Glasser, 2008, The study found sicker patients may present during off-hours because of inability to wait for routine hours. The researchers postulated that off-hour patients are USA. 82 sicker because MI onset while sleeping may lead to longer ischemic times upon arrival. The study found higher rates of cardiogenic shock and higher rates of multi-vessel coronary artery disease in those patients presenting during off-hours. However, after accounting for these factors, off-hour presentation remained associated with poorer outcomes. No Kuijsten, 2015, Researchers disclosed that although APACHE II was used to correct for illness severity patients were matched based upon propensity score, this did not fully Netherlands. 137 exclude the influence of case mix differences. The study found that surgical patients admitted in the middle of the night are different from patients admitted during office hours. The study found that there as an increase in the difference between observed and predicted mortality from 4pm onwards. No Laupland, 2011, Patients admitted during weekdays were different based on a number of characteristics from those admitted during nights and/ or weekends. This study identified Canada. 101 that while most admissions to ICU occur in the afterhours and that the weekends admissions were associated with higher crude case fatality, the day of the week or night time admissions were not associated with mortality once adjustment for confounding variables were performed. No Meynaar*, 2009, This study found no difference in case mix adjusted hospital mortality between ICU patients admitted during daytime as compared with those admitted during Netherland. 105 office hours. This study found no mortality difference in urgent surgery and in medical patients between patients admitted during daytime and patients admitted off-hours. Standardised mortality ratios are no different for patients admitted during daytime as compared to those admitted during off-hours. Logistic regression analysis confirmed that age APACHE II and expected mortality and admission type were related to outcome but off-hour admission was not. Results were similar with SAPS II expected mortality instead of APACHE II expected mortality. The researchers conclude that the differences in mortality could all be explained by differences in disease severity. No Morales*, 2003, When researchers implemented the logistic regression analysis model with hospital mortality rate as the dependent variable and admission source, admission USA. 78 diagnosis category, APACHE III predicted hospital mortality rate, and group of night time admission as independent variables, we did not find significant difference in mortality rates between the early and late night time admissions (p= 0.6191; OR: 1.063; 95% CI: 0.835–1.354). No Numa, 2006, After multivariate analysis, 3 of 9 diagnostic subgroups analysed (congenital heart disease, cardiac arrest, and shock) demonstrated a significantly higher risk of Australia. 120 death for evening but not weekend admissions. Mixed Uusaro*, 2003, The study findings were not explained by disease severity, intensity of care given to patients in ICU or by decisions to restrict patient’s care. Finland. 84 No Wunsch*, 2004, The researchers conclude that the finding that admission to an ICU at night and at the weekend is not associated with significantly higher case mix adjusted UK. 108 hospital mortality is reassuring. The association for night was still significant after adjustment for case mix using the UK APACHE II model (OR (night): 1.16; 95% CI: 1.10–1.23) but was non-significant when adjusted using the component model (APACHE II components OR (night): 1.02, 95% CI: 0.96–1.09). No Hixson, 2005, The study found that out of hours admissions were more likely to be emergency, non-operative patients; differ in the distribution of patients by primary diagnosis USA. 28 and have lower PRISM III score but have a higher overall PRISM III predicted mortality risk. The researchers purported that the reason for this has to do with the differences in the number of emergency, non-operative patients, cardiovascular surgery patients, and the frequency of cardiopulmonary resuscitation for patients admitted at the weekend and their effects on the PRISM III risk of mortality model. No Arias, 2004, Paediatric patients admitted during out of hours had a higher odd of death than did patients admitted during the daytime hours, after adjustment for the severity of USA.130 illness and other covariates diagnosis specific analyses (except for shock or congenital cardiovascular disease) did not demonstrate different mortality rates when out of hours admissions were compared with daytime admissions. No Arslankoylu, The study found the velocity of admissions to the PICU was higher in daytime and weekdays than evenings and weekends. That may be due to the delay at the 2008, Turkey. 67 admission of the patients to the PICU, because parents may prefer daytime or weekdays to admit their children to the hospitals or may be less vigilant to the 37 symptoms of the critically ill children in the evening time. Truly the higher PIM2 scores and longer LoS of the daytime admissions support that thought. 38 c. Intensive Care patients Difference Author, Year Comment Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project in outcome Country No Schmulewitz, The study did not adjust data for severity of illness on admission although the research did subsequently look for information on those admissions that had passed 2005, UK. 37 through HDU and ICU settings. d. Neuroscience patients Difference Author, Year Comment in outcome Country Yes Niewada, 2012, The study found patients were older, more likely to have AF or coronary heart disease, were more disabled prior to stroke and suffered more severe stroke, which Poland. 119 corresponds to more urgent medical seeking and shorter mean time from onset admission.
No Albright, 2009, Study found no significant differences in NIHSS scores, age, race or admission glucose levels when comparing the weekend and weekday groups. In addition, the USA. 19 study did not find a significant difference in IV T-PA treatment rates. While the study did not find the rate of weekend admissions for AIS- T-PA to be significantly different between the 2 sites (37%, 35%), it did find the weened stroke rate to be greater than the expected 28.6% 24/7. This increased incidence of stroke on weekends is consistent with previous reports of an increased incidence of stroke on holidays and weekends. No Almekhlafi, 2014, The study found baseline differences between the two groups – with atrial fibrillation and smoking being more prevalent in the cohort treated during evenings and Canada. 129 weekends. It was also expected that the two groups have other unknown clinical and pathophysiological differences. Yes Campbell, Patients admitted out of hours are more likely to present with haemorrhagic stroke, have reduced consciousness and have pre-morbid dependency; those 2014,UK. 68 patients with ischaemic stroke are more likely to present with more severe stroke subtypes. These data suggest that the observed excess mortality associated with out of hours admission reported in previous studies can largely be explained by un-measured differences in severity and prognosis. However, despite presenting with a greater illness severity, patients admitted out of hours are also less likely to receive timely access to key investigations and interventions, such as brain scanning and stroke unit admission. No Karlinski, 2013, The study found, patients admitted at out-of-office hours and on non-working days appear to be very similar to patients admitted during regular working hours, Poland. 135 both in terms of baseline characteristic and outcome. The tendency for a lower proportion of mild strokes may correspond with the higher proportion of the hyper dense artery sign, but it does not seem to modify the effect of thrombolysis. Yes Reeves, 2009, This study found clinically important differences in the study for ischemic and haemorrhagic stroke was non-existent for patients presenting off or on-hours. USA. 72 Mixed Sheu*, 2007, This study found that patients with AMI admitted during off-hours were more likely to have a STEMI, present with cardiogenic shock and develop shock after Taiwan. 33 presentation compared with those admitted during regular hours. After adjustment for the difference in risk, off-hours admission was not significantly associated with higher complication rates, especially ventricular arrhythmias and gastrointestinal bleeding. There was no difference in outcomes between STEMI and NSTEMI. The researchers found higher complication rates for patients admitted during off-hours. e. Surgical (Emergency) patients Difference Author, Year Comment in outcome Country No Buse, 2004, The study found that older age and higher severity of injury were predictive of increased major complications and mortality following trauma care. The study USA. 132 identified that for patients requiring emergency trauma surgery, major complications and mortality were not associated with whether a patient presented during the night or day, on a weekend v. weekday, month or presentation or year of presentation. No Orman, 2012, A fully adjusted model was created that controlled for age, race, sex, diabetes, BMI, vasopressor support, dialysis, previous upper abdominal surgery, portal vein USA. 31 thrombosis, re- transplantation, donor location, pre-MELD era v. MELD era, diagnosis, donor age, cold and warm ischemia times, night-time organ procurement, donor cause of death, and daytime operations v. night-time operations. The results for each of these models were unchanged from the results for the unadjusted model. f. Trauma patients Difference Author, Year Comment in outcome Country No Carmody, 2002, The study found more patients were admitted on weekends and at night-time and they had higher injury severity and a higher percentage of penetrating trauma. USA. 133 When patients were matched for injury severity the researchers found no evidence of increased mortality related to fixed trauma team staffing. No Laupland, 2009, The study identified that the case-mix of patients was an important determinant of outcome and variation during times of the day and days of the week and likely Canada. 70 explains much of the afterhours increased mortality effect seen. However, after controlling for diagnosis and severity of disease and case mix reduced (eliminated) unequal outcomes of patients. There researchers surmised that adjustment for case mix is clearly an important consideration to define consistency in processes of care afterhours. Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project
TABLE 34 (a–e): Data extraction of administrative studies identifying staffing as a contributory factor of predicting unequal health outcomes for patients admitted out of hours
a. Cardio Respiratory patients Difference Author, Year Comment in outcome Country Yes Aujesky, 2009, It is purported that the possible association between weekend admission and increased 30-day mortality (particularly among the most severely ill patients) is that USA. 73 fewer medical providers and professional staff tend to work in-hospitals on weekends than on weekdays and those who do work on weekends may have less clinical experience. Also, fewer supervisors are present on weekends and they are often responsible for supervising the work of staff members they do not know as well. Yes Deshmukh, Authors purport that one possible reason for the difference in outcome could be differences in staffing. During weekends, hospital staffing is reduced in overall 2012, USA. 80 quantity and in the number of qualified to perform certain procedures. It may be surmised that subtle early signs of acute problems go unnoticed until later. In many hospital settings, a physician not entirely familiar with the patients’ problem is likely to provide coverage on weekends; hence, the patient might not have access to a physician familiar with all the medical issues. b. Gastroenterology patients Difference Author, Year Comment in outcome Country Yes Button*, 2011, Researchers purport that possible explanation for the increased mortality at weekends and on public holidays include reduced staffing levels that may lead to less UK. 74 thorough assessment, lack of specialist or senior consultant cover, less application of multidisciplinary team care, poor communication at handover and possible delay to investigation such as endoscopy in some hospitals without out of hours services. Yes Dorn,* 2010, The study portends a worse outcome suggests that changes in the delivery of care (i.e., processes) have the potential to improve out-comes. Considering that USA. 43 those seeking care at the weekend are likely sicker, one possible strategy is to increase hospital staffing levels. c. Intensive Care patients Difference Author, Year Comment in outcome Country Yes Bhonagiri, The study identified that the unequal outcomes for patients admitted to out of hours may equally be related to patient selection for elective surgery, patient workup 2011, for elective surgery, surgical team skill mix or hospital resources. Finally, a lack of early assessment and formation of a management plan by intensivists may be the Australia. 111 reason for this difference. 39 40 d. Medical patients Difference Author, Year Comment
Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project in outcome Country Yes Ricciardi, Mortality following a weekend admission for patients admitted to hospital with resident trainees was significantly higher than hospitals with no resident trainees.The 2014, USA. 16 authors suggest that it may be a problem with supervision of trainees. e. Neuroscience patients Difference Author, Year Comment in outcome Country Mixed Bejot, 2013, The onset during weekends/ public holidays was associated with higher risk of 30-day mortality during 1985–2003 but not during 2004–2010; before and after the France. 81 introduction of a dedicated stroke care network. No Crowley, 2009, The researchers purported that there are fewer physicians and nurses in-house on weekends, with a concomitant increase in individual workload on those health USA. 75 care providers. Additionally, the physicians who are working on weekends are often less experienced than those available on weekdays, and they will often cover patients with whom they are less familiar. It might seem intuitive that this difference in experience would be more pronounced in academic facilities, where resident coverage is the norm. Yes Saposnik, 2007, Researchers proposed increased mortality with stroke admissions on weekends in rural settings compared to urban settings that competing tasks and the less Canada. 106 experience of the treating physicians in the rural setting, who may be general practitioners as opposed to stroke specialist, may be the cause of increased mortality of afterhours admissions.
TABLE 35 (a–g): Data extraction of clinical studies identifying staffing as a contributory factor of predicting unequal outcomes for patients admitted out of hours
a. Cardio Respiratory patients Difference Author, Year Comment in outcome Country Yes Glasser, 2008, Higher angioplasty failure rates occur at night, although there has not been sufficient angiographic or procedural information to understand potential causes. In USA. 82 this respect, those complications that may be related to operator performance, such as vessel dissection, were more frequent at night, whereas complications potentially related to lesion characteristics, such as distal embolisation, were not. The use of time-consuming devices was also less frequent at night than during daytime hours, despite similar lesion characteristics. No Jneid, 2008, The study found that STEMI patients arriving during off-hours were more likely to receive fibrinolytic therapy and slightly less likely to undergo primary PCI. Whilst USA. 59 this did not result in any net differences in overall reperfusion rates, however may reflect an appropriate clinical decision on the part of off-hour physicians to select fibrinolytic therapy when prompt PCI is not feasible. The researchers purport that regardless of the soundness of these triage decisions, the timeliness of primary PCI, when selected, was suboptimal, pointing out the ongoing system challenges in achieving rapid system activation and staff mobilisation during off-hours. Yes Lairez, 2009, The researchers purported that the results evolve from less efficient patient management during night-time hours.This study reflects actual clinical practice and France. 52 demonstrates the existence of greater mortality risk for emergency PCI performed at night than for daytime emergency PCI, independent of indication and previous treatment. The researchers conclude that better management of ACS is needed and, perhaps, that greater use of adjunctive medications should be considered when PCI is performed at night. No Sorita, 2014 The researchers purported that the mortality increase during off-hours may be alleviated with institutional systems improvement. USA. 83 b. Gastroenterology patients Difference Author, Year Comment in outcome Country Yes DeGroot*, 2012, Found proposed that physician related factors may be a factor but was unlikely because they did not identify an increased mortality at other times when physician Netherlands. 42 numbers were low (night and evening). b. Gastroenterology patients Difference Author, Year Comment in outcome Country No Haas*, 2010, The study found that there was no difference in outcome for admissions out of hours. The researchers surmised that the reason for this was that a competent USA. 27 support staff is available at all times as well as the ability of the emergency room physician, internist and gastroenterologist to appropriately risk strategy patients early in their course, validating the consensus recommendations for managing patients presenting with GI bleed . c. Intensive Care patients Difference Author, Year Comment in outcome Country No Arabi, 2006, The study demonstrates that a majority of patients are in fact admitted during the after-hours, a period in which they are in most need of a qualified intensivist. Saudi Arabia. 21 Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project Mixed Arias, 2004, ICU specific information regarding differences in staff and physician coverage was not present in the database and might independently influence mortality rates. USA. 130 No Arslankoylu, The study found that in day time admissions, mechanical ventilation initiation ratio was higher than the evening admissions. The study purports that the difference 2008, Turkey. 67 might be due to the presence of a paediatric intensivist in the PICU during daytime. Mixed Ensminger*, The study purported that the structure of their critical care practice compared to structure of ICUs investigated in other studies may be responsible for the absence 2004, USA. 26 of overall increase in the mortality of patients admitted to the ICU on weekends in the study. No Meynaar*, For the purposes of this study the definition of off-hour was based on the presence or absence of the intensivist in the ICU. 2009, Netherlands. 105 No Morales*, Results suggest that night-time admission to an intensive is unit need not be associated with poor out-come as long as adequate staffing and services are 2003, USA. 78 maintained. Provided that there are enough nurses, respiratory therapists, physicians, and other medical personnel as well as laboratory, radiology, and other services needed to provide optimal patient care, the timing of ICU admission is unlikely to be associated with mortality rate. Yes Neuraz, 2015, The risk of death increased when the patient-to-nurse ratio was greater than 2.5 and when the patient to physician ratio exceeded 14. The highest ratios occurred France. 79 more frequently during the weekend for nurse staffing and during the night for physicians. Mixed Sheu*, 2007, The study found that results were somewhat compatible with the staffing pattern in the ICU where resident physicians and critical care nurses staffed in the ICU Taiwan. 33 in hours at constant levels all the time, while dedicated intensivists led the morning rounds on all days of the week but did not stay in house overnight. The study found that there were fewer ICU admissions and non-conferences on weekends. Researchers purported that with the same staffing levels the lighter workload might result in better patient care and therefore, lower hospital mortality for patients admitted on weekends. Researchers observed that the adjusted OR of hospital mortality were higher for morning admissions than for afternoon admissions, and were higher for midnight admissions than for evening admissions implied that there might be ‘morning-effects’ or ‘sleep deprivation effects’ beyond the effect of staffing levels. The study found that routing morning service, teaching rounds, case conferences and journal clubs were all held in the morning. Mixed Uusaro*, 2003, Most of the ICUs in this study lacked dedicated full time intensivist coverage for the whole-time period from Friday afternoon till Monday morning. Researchers Finland. 84 found that mortality was not increased for patients who were admitted to ICU from 1600 hours to 0800 hours. The researchers speculate that of the ICUs included in the study the physician on-call in-hospital for critically ill patients most often does not work on a daily basis in the ICU during normal working hours. Therefore, the physician on-call does not necessarily have in-depth knowledge of the patients who already are in the ICU at the start of her/his shift (1600 hours). New admissions by this physician occur from 1600 hours to 0800 hours and therefore she or he has to know new patients in detail to make plans for examinations and therapeutic interventions. It may be that during the evenings and night time the physician is more informed of and involved with newly admitted patients and thus more dedicated to their care. This could translate into equal mortality for “out-of-office” hour and “office hour” admissions. No Wunsch*, 2004 Researchers found that patient care on admission to ICU does not appear to be affected by lower levels of either onsite staffing in the ICU, or of support services UK. 108 locating elsewhere in the hospital during certain days and times. d. Neuroscience patients Difference Author, Year Comment in outcome Country Mixed Bray, 2014, Researchers purported that the highest risk of death observed was in stroke services with the lowest nurse / bed ratios. 41 UK. 55 42 Yes Campbell, 2014, Researchers broke down stroke care into a number of bundles of which nursing ratios and senior consultant presence was analysed. Interestingly, they reported UK. 68 that 24/7 consultant ward rounds had no effect on stroke mortality with weekend admission but the nursing ratio did. They identified that a lower ratio of nurses to patients had a higher mortality. Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project Yes Hasegawa, The study purport that the staffing level on admission is also likely to be an important element of stroke unit care and researchers conclude that health policy 2005, Japan. 48 makers should be concerned about the staffing level on admission and the ratio during the early stages of stroke. e. Orthpaedics patients Difference Author, Year Comment in outcome Country Yes Foss, 2006, Staff reduction during holiday periods in units that care for acute surgical patients may adversely influence postoperative outcome. This may have important Netherlands. 77 consequences both for outcome analysis of interventions and the planning of resource management in surgical units. f. Trauma patients Difference Author, Year Comment in outcome Country No Carmody, The study found that no clinical significant diurnal-weekend or volume related variation in quality of care. Consistency of outcomes implies consistency in quality 2002, USA. 133 of care. The researchers concluded that if a centre is staffed appropriately an injured patient should have the same risk of mortality regardless of time or day of admission. No Laupland, 2009, Trauma staffing is lower in the afterhours as compared to weekday time hours with fewer resident physicians, nursing staff and allied health professionals such as Canada. 70 pharmacists, physiotherapists and respiratory therapists. The study purported that trauma surgeons are not mandated to remain in hours but frequently do during busy trauma seasons. g. Surgical patients Difference Author, Year Comment in outcome Country No Orman, 2012, At centers where the primary surgical team is on call for a week at a time or for the whole weekend, there may be a resultant decrease in graft survival later in the USA. 31 weekend preceding days. Another explanation for a decline in graft survival at 1-year could involve the support staff for surgical services. During the weekend, these individuals may form a cross-disciplinary team involved in a heterogeneous set of operative cases, including trauma, orthopaedics, and surgical emergencies. During weekdays, the surgical team is likely to be composed of dedicated transplant personnel. Procurement may be left to local surgical teams at non-liver transplant centers more often on weekends, whereas transplant teams may travel to the donors on weekdays and weeknights. Perioperative support, including the physical presence of multiple medical experts (e.g., dedicated transplant pharmacists), may be less available on weekends. Therefore, there are plausible reasons for weekend transplants to have worse graft survival.
TABLE 36 (a–d): Data extraction of administrative studies identifying treatment delay as a contributory factor of predicting unequal health outcomes for patients admitted out of hours
a. Cardio Respiratory patients Difference Author, Year Comment in outcome Country Yes Aujesky, 2009, Authors purport that understaffing in the emergency and radiology departments, numerically and in terms of expertise could potentially result in delayed diagnosis USA. 73 and treatment of PE, with an unfavourable impact on patient prognosis. Provision of care by covering physicians and / or more junior physicians may lead to the underuse of recommended processes of care for PE and that are associated with improved patient outcomes. Inadequate professional staffing and medical coverage during the weekend may also delay the detection of potentially fatal early complications. Yes Deshmukh, Patients admitted on weekends were less likely to undergo cardio version than those admitted on weekdays (7.9% v. 16.2%; p= 0.0001). Also, the interval between 2012, USA. 80 admission and performance of procedures was longer for patients admitted on weekends. When cardio version was added to the regression model, the difference in weekend and weekday mortality was not significant. a. Cardio Respiratory patients Difference Author, Year Comment in outcome Country Yes Kostis, 2007, The time between admission and performance of procedures was longer for patients admitted on weekends. The percentage of patients who underwent a USA. 50 procedure on the day of admission (possibly reflecting primary PCI) was also lower on weekends. Yes Kruth, 2008, Data show a markedly longer in-hospital delay for primary angioplasty during the weekends as compared to traditional working hours. However, within the observed Germany. 69 in-hospital delays door to angiography time did not influence in-hospital mortality. b. Gastroenterology patients Difference Author, Year Comment Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project in outcome Country Yes Button*, 2011, Recorded endoscopy was significantly lower (p<0.001) for admissions on Fridays (37.1%) and Saturdays (38.8%) than on other days (which varied between 44.3% UK. 74 for Sundays and 47.8% for Tuesdays). Rates of endoscopy on the day of admission were lower (p<0.001) for admissions on Saturdays (8.5%) and Sundays (7.4%) than on weekdays (17.5% to 20.9%). The median time to endoscopy was higher (p<0.001) for admissions on Fridays and Saturdays (both 3 days) and on Sundays (2 days) than for other days of the week. Yes Dorn*, 2010, The study found that patients admitted over the weekend were less likely to undergo endoscopy, when we controlled for EGD in our analysis the weekend mortality USA. 43 effect did not change. No Jairaith*, 2011, Here were significant delays to endoscopy in those patients who presented on weekends with only 38% receiving endoscopy within 24 hours of presentation. UK. 49 c. Intensive Care patients Difference Author, Year Comment in outcome Country Yes Bhonagiri, 2011, After-hours return to the ICU following elective surgery may imply prolonged surgery started in hours, where intraoperative complications have delayed ICU Australia. 111 admission, but the resulting physiological derangement is not reflected in theAPACHE III- risk of death. Alternatively, these admissions may be of patients whose elective surgical procedures were started at times when a lack of normal facilities, resources and staff have put the patients at increased risk of death. d. Neuroscience patients Difference Author, Year Comment in outcome Country No Albright, 2009, The study was unable to find a significant difference in IV t-PA treatment rates in patients presenting on a weekend and those presenting on a weekday (60.4 v. USA. 19 58.5%: p= 0.397). Mixed Karlinski, 2013, In comparison to patients admitted during regular hours, the proportions of those treated within 90 min from the stroke onset were lower in all evaluated groups. Poland. 135 Considering the similar rate of DNT 60 min, it may be assumed that increased delays were most likely due to prolonged pre-hospital phase. Yes Palmer, 2012, The researchers found that some of the measures that reveal a significant disparity, particularly same-day brain scan rates, are less likely to be affected by UK. 53 variations in stroke severity (unadjusted OR: 0.83 [95% CI: 0.81-0.86]) and thrombolysis (unadjusted OR: 0.82; 95% CI: 0.73–0.92)
TABLE 37 (a–e): Data extraction of clinical studies identifying treatment delay as a contributory factor of predicting unequal health outcomes for patients admitted out of hours
a. Cardio Respiratory patients Difference Author, Year Comment in outcome Country No Al-lawati, 2012, The study found similar proportion of patients in the weekdays group and weekend group received thrombolytic therapy within the recommended time period (≤ 30 43 UAE. 128 min) (32 v.27%; p= 0.25). Furthermore, recommended door-to-needle time (≤ 90 min) was not significantly different be-tween the 2 groups (51% in weekdays and 53% in weekends; p= 0.87). However, PCI (31 v. 26%; p= 0.006) and CABG (3.2 v. 1.7%; p= 0.01) were significantly higher in admission on weekdays compared with weekends. 44 b. Gastroenterology patients Difference Author, Year Comment Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project in outcome Country Yes DeGroot*, 2012, The study found that patients admitted during the evening had a significantly longer time to endoscopy compared with patients admitted during daytime and night Netherlands. 42 and 16 h respectively for day, evening and night, p<0.01). However, time between admission and endoscopy was not significantly different during the weekend than during the week (15.9 h v. 17.6 h). The need for endoscopic intervention was not different between the two groups. No Haas*, 2010, The study found that patients presenting with hematemesis received upper endoscopy earlier than patients presenting with melena. There was no statistically USA. 27 significant difference in the rate of surgical interventions was noted for weekend v. weekday admission (4% v. 1.61% p= 0.325). Researchers also purported the reason for no difference in outcome for patients admitted out of hours is that endoscopy for this site is available 24 hours a day, 7-days a week and that the standard of practice employed by the gastroenterology department is the same regardless of the day of the week. No Myers*, 2009, The mean (± SE) time to endoscopy was similar between groups (weekend 1.14 ± 0.03 days v. weekday 1.12 0.03 days; p= 0.60); however, patients admitted Canada. 91 at the weekend were less likely to undergo endoscopy on the day of admission (43% v. 45%; p= 0.01). This difference was most pronounced by the second day, at which time 81% of patients admitted on a week day had undergone endoscopy v. 75% of those admitted at the weekend (adjusted OR: 0.68; 95% CI: 0.63 to 0.73; p<0.0001). After the second day, there were several small but statistically significant differences in the timing of endoscopy between patients admitted at the weekend v. those admitted on a weekday. c. Intensive Care patients Difference Author, Year Comment in outcome Country Yes Glasser, 2008, The study found off-hour angioplasty was not associated with significant delay when using symptom onset to PCI time. Further researchers found that incidence USA. 82 of adverse events was only slightly higher on weekends, a time when delays would be expected to be greater than weekdays with the majority of difference consistently occurring between night time v. daytime PCI regardless of weekday or weekend. d. Neuroscience patients Difference Author, Year Comment in outcome Country Yes Niewada, 2012 The study found that although weekend patients were more likely to arrive within the first 3 or 6 hours after stroke onset, it did not result in more re-PA treatment Poland. 119 procedures.
No Almekhlafi, 2014, Significant delays in imaging to reperfusion times were encountered during evenings and weekends compared with working hours. Despite this, the study reported Canada. 129 imaging to reperfusion times remain relatively short compared with the existing literature. Yes Campbell, 2014, The study found significant differences in the quality of care received by patients admitted out of hours. Patients admitted out of hours waited longer to receive a UK. 68 brain scan or be admitted to a stroke unit and were less likely to be admitted to a stroke unit directly or to receive thrombolysis, multidisciplinary stroke specific care and therapy early after admission. Yes Reeves, 2009, Clinically important differences in the quality of care provided to patients who presented during off- or on-hours were small to non-existent. The proportion of USA. 72 patients who arrived within 2 hours who were treated with intravenous tissue plasminogen activator was slightly lower during off-hours (56.4% v. 58.8%), but deep vein thrombosis prophylaxis rates were slightly higher in the off-hour group (67.4% v. 65.6%). No Sato, 2015, The study found no evidence in heterogeneity in the effects according to the time of day for the initiation of treatment. The researchers found that intensive BP Australia. 36 lowering is likely to be a generalisable treatment strategy irrespective of admission hours among patients with acute ICH under assuring background stroke care. e. Orthopaedics patients Difference Author, Year Comment in outcome Country No Mathews, 2016, Time to surgery There was statistically no difference in the odds of time-to-surgery being less than 36 h between weekend and weekday patients (p= 0.975). As UK. 61 ASA increases by one unit, the expected odds of having a time-to-surgery of less than 36 hours are reduced by 32% (p= 0.001; 95% CI: 0.54–0.85) The season also has a significant effect on undergoing surgery within 36 hours (p= 0.014). Patients who were admitted in spring, summer or autumn all have greater odds of having a time to surgery of less than 36 hours compared to the patients admitted in winter. Chapter 7: Which care processes have been effective in reducing unequal health outcomes for patients admitted out of hours and at the weekend?
It has been suggested that patients presenting out of hours are subject to inferior care. Of the 146 studies, we have examined in more detail an attempt has been made to investigate the correlation between care received by patients admitted out of hours.
A specific model of care was defined as one in which a patient with a particular medical condition is assessed, treated, and cared for by a team of health professionals experienced in treating that condition. This team is responsible for the patient’s care on admission to hospital and throughout their hospital stay ensuring a continuous level of care is provided. The team includes a senior medical specialist, appropriate and consistent nursing ratios, experienced allied health members and access to investigations and treatments in a timely fashion.
For example, a patient who presents with a stroke is assessed, treated and cared for in a stroke unit. This stroke unit includes stroke specialists, nurses and allied health professionals that are experienced in treating stroke patients, and has access to the investigations and treatments that are required for patients presenting with a stroke.
Other units displaying specific models of care identified in the review include: the trauma unit that manages patients presenting following trauma; the cardiology unit that is responsible for performing percutaneous coronary interventions in patients presenting with acute myocardial infarct; and the intensive care unit that manages the most critically unwell patients, H&M unit that accepted direct admissions for H&M.
Thirty-two studies analysed a specific model of care. Fourteen studies analysed ICUs, eight comprehensive stroke centres, four percutaneous coronary intervention clinics, four trauma care centres, one transplant team and one H&M unit. The characteristics and findings of the included studies are summarised in Tables 36–42.
Studies whose methodologies were not restricted to specific models of care were not included. Studies excluded were investigations of patients admitted to hospital under units with inconsistent staffing or with varying levels of experience with stroke. For example, the methodology used by Saposnik (2007) 125 was one that included patients suffering from a stroke who presented to acute care facility across Canada. These facilities varied between rural, urban and tertiary level hospitals, and as such the patients were cared for in units that may not have been a dedicated stroke unit, instead been staffed by medical and nursing teams with less experience with stroke care.
Four 19 21 84 100 papers were prospective observational studies with only two studies 71 81 performed a pre and post investigation of the model of care or used a control group involved.
The majority of studies (18) 19-21 27-29 31 36 67 70 78 83 103 105 108 129 132–133 did not identify an association between out of hours admission and mortality. This was consistent across all different units providing a specific model of care. This is most evident in the studies on ICU, stroke and trauma admissions. Interestingly the studies in cardiology looking at PCI at first appear to be mixed. However, further analysis reveals that the studies reporting increase in mortality for patients admitted out of hours are older and may not reflect current 24/7 usage of PCI. With regards to GI bleeding the one 27 study identified that where patients were admitted and treated by a specific unit that deals with GI bleeding found no difference in outcome.
It cannot be inferred from these studies that specific models of care ameliorate the unequal outcomes for patients admitted out of hours. From the information reviewed there is no way to determine if there was a problem with out of hours admission prior to the introduction of the models of care or if there has been any improvement in mortality after the introduction of the model of care.
Even if such a connection could be made, the units providing the specific models of care are made up of a number of components and further investigation is required to identify which of components of the team improve outcome. While we have found no evidence base on how to best organise and operate a hospital out of hours to ensure there is no effect on mortality, what we have found is that patients admitted to and treated by a specific health care team did not have an increased mortality if they were admitted out of hours.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 45 46 Comprehensive Stroke Centres
Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project TABLE 38: Data extraction of studies analysing specific model – Comprehensive Stroke Centre/s
Author Comment Albright, 2009, USA. 19 Stroke Centre may ameliorate differential outcomes in stroke patients. These results may be due to the 24/7 availability of stroke specialists, advanced neuroimaging or ongoing training & surveillance of specialised nursing care. Results require confirmation in prospective studies. Albright, 2012, USA. 20 No evidence of differences in outcome. Ischemic stroke patients admitted on weekends did not display significant difference in LoS, inhospitable mortality, favourable hospital discharge or functional outcome at discharge or an inferior 90-day functional outcome or higher 90-day mortality. Almekhlafi, 2014, Canada. 129 Some delays were uncounted during evenings and weekend hours. Despite that, it was feasible to achieve relative short image to treatment during these hours in comparison to existing literature (no analysis of mortality just time to intervention after hours compared to in hours in dedicated stroke centre). Bejot, 2013, France. 81 The deleterious effect of weekends and holidays on 30-day mortality disappeared after the organisation of a dedicated stroke care network in the community. Hasegawa, 2005, Japan.48 Although specific elements of structures and processes for the effects of stroke unit care have not yet been identified, several possible factors have been emphasised, i.e. a multidisciplinary team approach, well-trained nursing team, early mobilisation, early hydration, early prescription of aspirin for ischemic stroke and protocol adherence. The staffing level on admission is also likely to be an important element of stroke unit care. Karlinski, 2013, Poland. 135 There is no bad time for thrombolysis. Stroke centres should feel confident about treatment out of regular working hours, irrespective of equipment and staff availability. However, it may be reasonable to pay attention during night-time. McKinney, 2011, USA. 71 More appropriate hospital staffing and organisation of stroke care such as that provided by CSC may negate differences in outcome and save lives. Sato, 2015, Australia. 36 Off-hour admission was not associated with poor outcome in patients with acute ICH who participated in the trial. Disadvantage of off-hour admissions may be avoidable when stroke care is standardised and based in 24-hour monitored facilities.
Intensive Care Units
TABLE 39: Data extraction of studies analysing specific model – Intensive Care Unit/s
Author Comment Arabi, 2006,Saudi Arabia. 21 The outcomes are similar for patients admitted during weekdays, weeknights and weekends to and ICU covered on site by qualified intensivist 24 hours a day 7-days a week. Arslankoylu, 2008, Turkey. 67 Although there may be some differences according to the time of admission to the PICU with respect of the overall mortality rates, the researchers conclude that closed PICU management under the control of a paediatric intensivist carries gratifying results in terms of continuousness and constancy of the PICU care. Bhonagiri, 2011, Australia. 111 Patients admitted to the ICU after hours have increase mortality and SMR. The increased mortality with after hours and weekend admission to the ICU is predominantly accounted for by patients with planned admissions following elective surgery. Ensminger* 2004, USA. 26 No significant differential outcome for medical or multispecialty patients. Hixson, 2005, USA. 28 Found no difference in mortality associated with weekend v. weekday admission in paediatric ICUs or the paediatric emergency department. Kuijsten, 2015, Netherlands. 137 ICU performance is influenced by an intricate interplay of various factors.The ICU organisation during off-hours there is intensive interaction with other medical disciplines, and changes in their quality of care during off-hours might influence ICU outcome as well, which is not reflected by illness severity at admission. The performance of health care workers (physicians and nurses) varies during the day. Although speculative, the detrimental effect of the circadian biorhythm on human performance during the night shift and especially at the end of the night shift is a known factor. Author Comment Laupland, 2011, Canada. 101 No association was found between timing of admission to ICU and subsequent outcome after controlling for a number of variables. Further investigation is needed to examine whether minimisation of after hour discharges and / or augmentation of ward care post ICU discharge may improve the ultimate outcome of critical illness. Lee*, 2008, Singapore 102 In a specialised ICU with adequate staffing and necessary diagnostic and therapeutic modalities, timing of severe head injury ICU admission is unlikely to be associated with mortality rate. Luyt*, 2007, France. 103 ICU off-hour admission is not associated with a poorer outcome than day shift admission consistent with a constant level of care, regardless of the time and day. Meynaar*, 2009, Netherlands. 105 We speculate that if intensivists are continuously present in the ICU during daytime and present when necessary during off-hours this is sufficient to avoid a quality
Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project gap during off-hours. Morales*, 2003, USA. 78 Night time admissions to ICU was not associated with a higher mortality rate or a longer hospital or intensive care unit stay compared with day time admission. Sheu*, 2007, Taiwan. 33 Non-office hours v. office hours admissions were not associated with poorer ICU, hospital or ventilator outcomes in a medical ICU equipped with patient management guidelines and staffed by intensives on call for 24 hours, who led the morning rounds on all days of the week but did not stay in hours over night. Moreover, time of day and day of week admissions to ICU were not associated with significant differences in-hospital mortality. Uusaro 2003, Finland. 84 In the ICU, the risk of dying is higher in evenings and during night-time as compared with daytime. When patients are discharged from the ICU, their risk of hospital death is not associated with the timing of discharge. If our findings can be confirmed elsewhere, they may have important implications for organisation of ICU services. Wunsch*, 2004 UK. 108 Admission to an ICU at night and at the weekend is not associated with significantly higher case mix adjusted hospital mortality.
Percutaneous Coronary Intervention and/or Critical Care Units
TABLE 40: Data extraction of studies analysing specific model – Percutaneous Coronary Intervention &/or Critical Care Unit/s
Author Comment Glasser, 2008, USA. 82 Finding appear related to both diurnal differences in presentation and lesion characteristics as well as differences in procedural complication and success rates that extend beyond differences in symptom to balloon time. Kruth, 2008, Germany. 69 Patients with STEMI admitted to hospitals with catheterisation facilities, admission during the off-hours is associated with higher in-hospital mortality. This may be due to lower rates of revascularisation therapy and longer pre-hospital and in-hospital delays compared to off-hours. Lairez, 2009, France. 52 The study reflects actual clinical practice and demonstrates the existence of greater mortality risk for emergency PCI performed at night than for daytime emergency PCI, independent of indication and previous treatment. The study shows that better management of ACS is needed and, perhaps, that greater use of adjunctive medications should be considered when PCI is performed at night. Sorita, 2014 USA. 83 Level I and Level II trauma centres were associated with smaller mortality increase during out of hours. The researchers purport that it is possible to achieve comparable outcomes during off-hours and not by simply adding more hospital staff to work during off-hours. 47 48 Trauma Clinics
Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project TABLE 41: Data extraction of studies analysing specific model – Trauma Clinic/s
Author Comment Busi ,2004, USA. 132 Study suggested that for patients requiring emergency surgery, major complication and mortality are not associated with whether patients presented during out of hours or not and that trauma units have long been aware of these issues, and may have made conscious efforts to maintain consistent standards of care at all times. Carmody, 2002, USA. 133 Researchers purport that it is essential for services providing continuous lifesaving care - notably trauma emergency medicine and critical care - to show that they have no clinically significant diurnal weekend or volume related variation in quality of care. Laupland, 2009,Canada. 70 No influence of the outcome of major trauma patients admitted to our large regional acute care hospitals. While inconsistencies in care in the afterhours may affect outcome of other patient populations, this does not appear to be the case for victims of major trauma, and may reflect the advance organisation of trauma care. Mitra, 2014, Australia. 118 In major trauma centres, equipment and tangible resource availability remains the same regardless of time of day. However, staffing levels vary, and there has been substantial debate about the importance of fully staffed trauma centres at all hours - including emergency physicians, trauma surgeons, anaesthetists, intensive care physicians, radiologists and broader support staff.
Transplant Teams
TABLE 42: Data extraction of studies analysing specific model – Transplant Team/s
Author Comment Orman, 2012, USA. 31 The outcome of this study provides reassurance about outcomes to the transplant community and patients awaiting transplantation. It highlights patient safety mechanisms, that is the availability of the appropriate number of surgeons and ancillary staff outside of hours.
Haematemesis and Melaena Units
TABLE 43: Data extraction of studies analysing specific model – Haematemesis & Melaena Units
Author Comment Haas*, 2010, USA. 27 The differences in outcome may be only a small part of the equation leading to adverse outcomes in patients presenting with UGIB but it remains a modifiable risk factors. The practice of this institution proves differences in outcome can be avoided in patients presenting with UGIB. Efficient and safe care should be implemented regardless of the day of admission. Chapter 8: Reflections, Implications and Conclusions
Reflections on the literature There are a number of limitations of the largely descriptive and retrospective studies included in this review making definitive recommendations from the literature difficult to make. The review identified that out of hours hospital admissions literature is bountiful and associations have been documented from an array of medical conditions, however it is unclear whether these associations are causal for a lower quality of care or are truly attributable to unobservable differences between patients. The papers that have been included have provided guidance and information to inform the next steps. In part, the questions that guided the review remain unanswered highlighting a number of gaps in our current knowledge. No studies were identified that specifically targeted the mechanisms to address unequal outcomes as a result of out of hours admission. However, a wider range of care processes were hypothesised and potentially contribute to our understanding of variation in the quality of care across the week. Future resourcing decisions to address unequal outcomes, from out of hours admission, requires further investigation of causation for the factors or combinations of factors that may contribute to adverse outcomes. Key findings There were essentially no papers found as a result of this rapid review that specifically examined the impact that interventions may have had if a differential outcome had previously been identified, for example in the areas of timeliness of senior decision making, timeliness to investigations, treatment time, models of care or staffing profiles.
In general terms, the weight of evidence that a differential outcome may exist is based largely on retrospective analysis of large administrative data sets, including those studying emergency department admissions out of hours.
Differential outcomes out of hours in clinical data sets, which were risk adjusted for severity of illness, were harder to demonstrate out of hours. However, it is important to note that illness acuity and severity is often reported as being higher out of hours in many of the reviewed articles, to account for higher unadjusted mortality, which is an important consideration in itself.
The vast majority of literature investigating outcome differences for patients admitted out of hours report against mortality (which is variably defined in terms of days after admission) and fails to consider other relevant outcomes. Examples include LoS, patient experience, readmission rates and other measures of patient morbidity such as healthcare associated adverse events.
Human resource, system or process factors that may be associated with or mitigate against a differential outcome has been identified in the review, but in general were hypothesised through expert analysis when the authors discussed their findings, rather than formally testing them to demonstrate causal linkage.
Unit based care delivery for specific cohorts of patients (e.g. ICUs PCIs and/or CCUs,Trauma Clinics, H & M and/or Transplant Teams) was associated with similar outcomes, irrespective of when patients were admitted, because resources were similarly allocated across all time periods. Next steps In the Queensland context, data linkage provides the opportunity to investigate if differences in outcomes exist locally, it’s magnitude, and if present whether this is a system wide phenomena, or demonstrates regional or diagnostic category variation. Any limitations of existing administrative studies were identified by the reviewers and will be incorporated into the analysis by CSIRO, where possible.
Specifically, system wide staffing data, cost data, patient experience surveys, and the death registry data will be linked to patient episode data to determine what individual factors or combination of factors could explain any discoverable differences in outcome.
As such, whilst mortality remains an important quantitative patient outcome differential, any demonstrable differential outcome will be analysed in the context of patient experience, staffing profiles and other patient outcome measures that extend beyond mortality.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 49 It is unrealistic to expect that out of hours and weekends can replicate weekdays in all facets of care and as such any improvement efforts, for identified outcome variation, are best targeted to regional or diagnostic variation. Given the contribution to overall health budgets comprised by clinical staffing, and its logical connection to care delivery out of hours, this is will be a particular area of focus for the research team.
Cost effectiveness analyses can then be undertaken to assist policy makers and health services executives to target areas of identified need where a differential outcome exists. Following on, high quality health services research is then required to demonstrate the impact of any intervention, so implemented, to ensure that clinical outcomes did in fact improve and that allocation of new or redistributed resources had been effective. The rollout of integrated electronic medical records confers the very real possibility that objective clinical and physiological admission data will enable future prospective research in this area.
Over time, such an approach will inform future planning, and provide the ability to develop services on the weekend and out of hours that are reflective of demonstrated areas of need. Strategic planning of hospital services can successfully overcome the variations in care that have been noted in acute admissions for some conditions. Acknowledgements
The authors of this report have been wholly responsible for all data collection, analysis and interpretation and for writing up their work. The Executive Committee, the REACH Project Steering Committee and Research Advisory have tried to ensure the accuracy of the authors’ work. Contribution of authors
Ms Michele Romeo, Principal Project Officer, Metro North Hospital and Health Services (Co-ordinating Investigator) undertook the proposal writing, designed and ran the literature searches, undertook reference sifting and evidence synthesis for the review, data extraction into summary tables; contribution to the writing of the report, including the discussion and conclusions; and managed the project.
Dr Roger O’Gorman, Medical Officer, Department of Emergency Medicine, Royal Brisbane and Women’s Hospital undertook evidence synthesis of the review, contribution of the writing of the report, including the discussion and conclusions.
Dr Anthony Bell, SOM UQ & Medical Director, Department of Emergency Medicine, Royal Brisbane and Women’s Hospital (Chief Investigator) contributed to the proposal writing contribution of the writing of the report including the discussion and conclusions. Data sharing statement
No new data have been created.
50 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project References
1. Mangold WD. Neonatal mortality by the day of the week in the 1974-75: Arkansas live birth cohort. American Journal of Public Health 1981;71(6):601-05. 2. Macfarlane A. Variations in number of births and perinatal mortality by day of week in England and Wales. BMJ 1978;2(6153):1670-73. 3. Mathers C. Births and perinatal deaths in Australia: Variations by day of week. Journal of Epidemiology and Community Health 1983;37(1):57-62. 4. Bell CM, Redelmeier DA. Mortality among patients admitted to hospitals on weekends as compared with weekdays. New England Journal of Medicine 2001;345(9):663-68. 5. Dixon-Woods M. Using framework-based synthesis for conducting reviews of qualitative studies. BMC Medicine 2011;9 (39;doi: 10.1186/1741-9-36). 6. Aylin P, Yunus A, Bottle A, et al. Weekend mortality for emergency admissions. A large, multicentre study. Quality and Safety in Health Care 2010;19(3):213-17. 7. Aylin P, Alexandrescu R, Jen M, et al. Day of week of procedure and 30-day mortality for elective surgery: Retrospective analysis of hospital episode statistics. BMJ (Clinical Research ed) 2013;346 f:2424. doi:10.1136/bmj.f2424 8. Concha OP, Gallego B, Hillman K, et al. Do variations in-hospital mortality patterns after weekend admission reflect reduced quality of care or different patient cohorts? A population-based study. BMJ Qualityand Safety 2013; bmjqs-2013-002218 9. Cram P, Hillis SL, Barnett M, et al. Effects of weekend admission and hospital teaching status on in-hospital mortality. The American Journal of Medicine 2004;117(3):151-57. 10. Freemantle N, Richardson M, Wood J, et al. Weekend hospitalization and additional risk of death: an analysis of inpatient data. Journal of the Royal Society of Medicine 2012;105(2):74-84. 11. Freemantle N, Ray D, McNulty D, et al. Increased mortality associated with weekend hospital admission: A case for expanded seven day services? BMJ (Online) 2015;351:h4596.doi:10.1136/bmj.h4596 12. Handel AE, Patel SV, Skingsley A, et al. Weekend admissions as an independent predictor of mortality: An analysis of Scottish hospital admissions. BMJ Open 2012;2(6):001789. 13. James MT, Wald R, Bell CM, et al. Weekend hospital admission, acute kidney injury, and mortality. Journal of the American Society of Nephrology 2010;21(5):845-51. 14. Mohammed MA, Sidhu KS, Rudge G, et al. Weekend admission to hospital has a higher risk of death in the elective setting than in the emergency setting: A retrospective database study of national health service hospitals in England. BMC Health Services Research 2012;12.(87). doi: 10.1186/1472-6963-12-87 15. Ricciardi R, Roberts PL, Read TE, et al. Mortality rate after nonelective hospital admission. Archives of Surgery 2011;146(5):545-51. 16. Ricciardi R, Nelson J, Roberts PL, et al. Is the presence of medical trainees associated with increased mortality with weekend admission? BMC Medical Education 2014;14(4). doi: 10.1186/1472-6920-14-14 17. Ruiz M, Bottle A, Aylin PP. Exploring the impact of consultants’ experience on hospital mortality by day of the week: a retrospective analysis of hospital episode statistics. BMJ Quality & Safety 2016;25:337-844.doi: 10.1136/ bmjqs-2015-004105 18. Pasupathy D, Wood AM, Pell JP, et al. Time of birth and risk of neonatal death at term: retrospective cohort study. BMJ (Clinical Research ed) 2010;341:c3498.doi:10.1136/bmj.c3498 19. Albright KC, Raman R, Ernstrom K, et al. Can comprehensive stroke centers erase the ‘weekend effect’? Cerebrovascular Diseases 2009;27(2):107-13. 20. Albright KC, Savitz SI, Ernstrom K, et al. Comprehensive stroke centers & the weekend effect: the SPOTRIAS experience. Stroke 2011;42(3):7. 21. Arabi Y, Alshimemeri A, Taher S. Weekend and weeknight admissions have the same outcome of weekday admissions to an intensive care unit with onsite intensivist coverage. Critical Care Medicine 2006;34(3):605-11. 22. Becker DJ. Do hospitals provide lower quality care on weekends? Health Services Research 2007;42(4):1589-612. 23. Cavallazzi R, Marik PE, Hirani A, et al. Association between time of admission to the ICU and mortality: A systematic review and metaanalysis. CHEST Journal 2010;138(1):68-75.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 51 24. Clarke MS, Wills R, Bowman RV, et al. Exploratory study of the ‘weekend effect’ for acute medical admissions to public hospitals in Queensland, Australia. Internal Medicine Journal 2010;40(11):777-83. 25. de Cordova PB, Phibbs CS, Bartel AP, et al. Twenty‐four/seven: a mixed‐method systematic review of the off‐shift literature. Journal of Advanced Nursing 2012;68(7):1454-68. 26. Ensminger SA, Morales IJ, Peters SG, et al. The hospital mortality of patients admitted to the ICU on weekends. CHEST Journal 2004;126(4):1292-98. 27. Haas J, Gundrum J, Rathgaber S. Comparison of time to endoscopy and outcome between weekend and weekday admissions to a community hospital in patients with upper gastrointestinal hemorrhage. American Journal of Gastroenterology 2010;105:s397-98. 28. Hixson ED, Davis S, Morris S, et al. Do weekends or evenings matter in a pediatric intensive care unit? Pediatric Critical Care Medicine 2005;6(5):523-30. 29. Lee KK, Ng I, Ang BT. Outcome of severe head injured patients admitted to intensive care during weekday shifts compared to nights and weekends. Annals-Academy of Medicine Singapore 2008;37(5):390. 30. Myers RP, Papay KD, Shaheen AAM, et al. Relationship between hospital volume and outcomes of esophageal variceal bleeding in the United States. Clinical Gastroenterology and Hepatology 2008;6(7):789-98. 31. Orman ES, Hayashi PH, Dellon ES, et al. Impact of night time and weekend liver transplants on graft and patient outcomes. Liver Transplantation 2012;18(5):558-65. 32. Ruiz M, Bottle R, Aylin P. The Global Comparators Project: International comparison of 30-day In-hospital mortality by day of the week. BMJ Quality & Safety 2015;24: 492-504. 33. Sheu CC, Tsai JR, Hung JY, et al. Admission time and outcomes of patients in a medical intensive care unit. Kaohsiung Journal of Medical Sciences 2007;23(8):395-404. 34. Magid DJ, Wang Y, Herrin J, et al. Relationship between time of day, day of week, timeliness of reperfusion, and in- hospital mortality for patients with acute ST-segment elevation myocardial infarction. JAMA 2005;294(7):803-12. 35. Worni M, Østbye T, Gandhi M, et al. Laparoscopic appendectomy outcomes on the weekend and during the week are no different: A national study of 151,774 patients. World Journal of Surgery 2012;36(7):1527-33. 36. Sato S, Arima H, Heeley E, et al. Off-Hour admission and outcomes in patients with acute intracerebral hemorrhage in the INTERACT2 trial. Cerebrovascular Diseases 2015;40(3-4):114-20. 37. Schmulewitz L, Proudfoot A, Bell D. The impact of weekends on outcome for emergency patients. Clinical Medicine 2005;5(6):621-5. 38. Abougergi MS, Travis AC, Saltzman JR. Impact of day of admission on mortality and other outcomes in upper GI hemorrhage: a nationwide analysis. Gastrointestinal Endoscopy 2014;80(2):228-35. 39. Ananthakrishnan A, McGinley E, Saeian K. Outcomes of weekend admissions for upper gastrointestinal hemorrhage: A nationwide analysis. Clinical Gastroenterology and Hepatology 2009;7(3):257-8. 40. Bejanyan N, Fu AZ, Lazaryan A, et al. Impact of weekend admissions on quality of care and outcomes in patients with acute myeloid leukemia. Cancer 2010;116(15):3614-20. 41. Byun SJ, Kim SU, Park JY, et al. Acute variceal hemorrhage in patients with liver cirrhosis: weekend versus weekday admissions. Yonsei Medical Journal 2012;53(2):318-27. 42. De Groot NL, Bosman JH, Siersema PD, et al. Admission time is associated with outcome of upper gastrointestinal bleeding: Results of a multicentre prospective cohort study. Alimentary Pharmacology and Therapeutics 2012;36(5):477- 84. 43. Dorn SD, Shah ND, Berg BP, et al. Effect of weekend hospital admission on gastrointestinal hemorrhage outcomes. Digestive Diseases and Sciences 2010;55(6):1658-66. 44. Gyenes GT, Yan AT, Tan M, et al. Use and timing of coronary angiography and associated in-hospital outcomes in Canadian non–ST-segment elevation myocardial infarction patients: insights from the Canadian Global Registry of Acute Coronary Events. Canadian Journal of Cardiology 2013;29(11):1429-35. 45. Hinds A, Ahmad D, Lopez K, et al. Weekday versus weekend admissions for upper gastrointestinal bleeding: Is there a true difference in patient outcomes? A meta-analysis. American Journal of Gastroenterology 2013;108:s479. 46. Fang J, Saposnik G, Silver FL, et al. Association between weekend hospital presentation and stroke fatality. Neurology 2010;75(18):1589-96. 47. Goodman EK, Reilly AF, Fisher BT, et al. Association of weekend admission with hospital length of stay, time to chemotherapy, and risk for respiratory failure in pediatric patients with newly diagnosed leukemia at freestanding US children’s hospitals. JAMA Pediatrics 2014;168(10): 925-31. 48. Hasegawa Y, Yoneda Y, Okuda S, et al. The effect of weekends and holidays on stroke outcome in acute stroke units. Cerebrovascular Diseases 2005;20(5):325-31.
52 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 49. Jairath V, Logan RF, Hearnshaw S, et al. Mortality from acute upper gastro-intestinal bleeding in the UK - Does it display a “weekend effect”? Gastroenterology 2010;138(5):s21. 50. Kostis WJ, Demissie K, Marcella SW, et al. Weekend versus weekday admission and mortality from myocardial infarction. The New England Journal of Medicine 2007;356(11):1099-109. 51. Kristiansen NS, Kristensen PK, Nørgård BM, et al. Off-hours admission and quality of hip fracture care: A nationwide cohort study of performance measures and 30-day mortality. International Journal for Quality in Health Care 2016;28(3):324-31. 52. Lairez O, Roncalli J, Carrié D, et al. Relationship between time of day, day of the week and in-hospital mortality in patients undergoing emergency Percutaneous Coronary Intervention. Archives of Cardiovascular Diseases 2009;102(12):811-20. 53. Palmer WL, Bottle A, Davie C, et al. Dying for the weekend: A retrospective cohort study on the association between day of hospital presentation and the quality and safety of stroke care. Archives of Neurology 2012;69(10):1296-302. 54. Shaheen AA, Kaplan GG, Myers RP. Weekend versus weekday admission and mortality from gastrointestinal hemorrhage caused by peptic ulcer disease. Clinical Gastroenterology and Hepatology: The Official Clinical Practice Journal of the American Gastroenterological Association 2009;7(3):303-10. 55. Bray BD, Ayis S, Campbell J, et al. Associations between stroke mortality and weekend working by stroke specialist physicians and registered nurses: Prospective multicentre cohort study. PLOS Med 2014;11(8): e1001705 56. Carr BG, Reilly PM, Schwab CW, et al. Weekend and night outcomes in a statewide trauma system. Archives of Surgery 2011;146(7):810-7. 57. Daugaard CL, Jorgensen HL, Riis T, et al. Is mortality after hip fracture associated with surgical delay or admission during weekends and public holidays? A retrospective study of 38,020 patients. Acta Orthopaedica 2012;83(6):609-13. 58. Inoue T, Fushimi K. Weekend versus Weekday Admission and In-Hospital Mortality from Ischemic Stroke in Japan. Journal of Stroke and Cerebrovascular Diseases 2015;24(12): 2787-92. 59. Jneid H, Fonarow GC, Cannon CP, et al. Impact of time of presentation on the care and outcomes of acute myocardial infarction. Circulation 2008;117(19):2502-9. 60. Kazley AS, Hillman DG, Johnston KC, et al. Hospital care for patients experiencing weekend vs weekday stroke: A comparison of quality and aggressiveness of care. Archives of Neurology 2010;67(1):39-44. 61. Mathews JA, Vindlacheruvu M, Khanduja V. Is there a weekend effect in hip fracture patients presenting to a United Kingdom teaching hospital? World Journal of Orthopedics. 2016;7(10):678. 62. Matsui K, Kojima S, Sakamoto T, et al. Weekend onset of acute myocardial infarction does not have a negative impact on outcome in Japan. Circulation: Official Journal of the Japanese Circulation Society 2007;71(12):1841-4. 63. Noman A, Ahmed JM, Spyridopoulos I, et al. Mortality outcome of out-of-hours primary Percutaneous Coronary Intervention in the current era. European Heart Journal 2012;33(34):3046-53. 64. Nandyala SV, Marquez-Lara A, Fineberg SJ, et al. Comparison of perioperative outcomes and cost of spinal fusion for cervical trauma: weekday versus weekend admissions. Spine 2013;38(25):2178-83. 65. Showkathali R, Davies JR, Kelly PA, et al. Admission at nights or weekends has no adverse effect on mortality for ST elevation myocardial infarction patients treated by primary Percutaneous Coronary Intervention. Single Centre Study from a large tertiary Cardiac Unit in United Kingdom.Journal of the American college of Cardiology 2013;60(17):136-7. 66. Rathod KS, Jones DA, Gallagher SM, et al. Out-of-hours primary percutaneous coronary intervention for ST-elevation myocardial infarction is not associated with excess mortality: A study of 3347 patients treated in an integrated cardiac network. BMJ Open 2013;3(6):e003063 67. Arslankoylu AE, Bayrakci B, Oymak Y. Admission time and mortality rates. The Indian Journal of Pediatrics 2008;75(7):691-4. 68. Campbell JT, Bray BD, Hoffman AM, et al. The effect of out of hours’ presentation with acute stroke on processes of care and outcomes: analysis of data from the Stroke Improvement National Audit Programme (SINAP). PLoS One 2014;9(2): e8746. doi:10.1371/journal.pone.0087946 69. Kruth P, Zeymer U, Gitt A, et al. Influence of presentation at the weekend on treatment and outcome in ST-elevation myocardial infarction in-hospitals with catheterization laboratories. Clinical Research in Cardiology: Official Journal of the German Cardiac Society 2008;97(10):742-7. 70. Laupland KB, Ball CG, Kirkpatrick AW. Hospital mortality among major trauma victims admitted on weekends and evenings: A cohort study. Journal of Trauma Management & Outcomes 2009;3:8. 71. McKinney JS, Deng Y, Kasner SE, et al. Comprehensive stroke centers overcome the weekend versus weekday gap in stroke treatment and mortality. Stroke 2011;42(9):2403-9. 72. Reeves MJ, Smith E, Fonarow G, et al. Off-hour admission and in-hospital stroke case fatality in the Get with the Guidelines-Stroke Program. Stroke 2009;40(2):569-76.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 53 73. Aujesky D, Jiménez D, Mor MK, et al. Weekend versus weekday admission and mortality after acute pulmonary embolism. Circulation 2009;119(7):962-8. 74. Button L, Roberts S, Evans P, et al. Hospitalized incidence and case fatality for upper gastrointestinal bleeding from 1999 to 2007: A record linkage study. Alimentary Pharmacology & Therapeutics 2011;33(1):64-76. 75. Crowley RW, Yeoh HK, Stukenborg GJ, et al. Influence of weekend versus weekday hospital admission on mortality following subarachnoid hemorrhage: Clinical article. Journal of Neurosurgery 2009;111(1):60-6. 76. Crowley RW, Yeoh HK, Stukenborg GJ, et al. Influence of weekend hospital admission on short-term mortality after intracerebral hemorrhage. Stroke 2009;40(7):2387-92. 77. Foss NB, Kehlet H. Short-term mortality in hip fracture patients admitted during weekends and holidays. British Journal of Anaesthesia 2006;96(4):450-4. 78. Morales IJ, Peters SG, Afessa B. Hospital mortality rate and length of stay in patients admitted at night to the intensive care unit. Critical Care Medicine 2003;31(3):858-63. 79. Neuraz A, Guérin C, Payet C, et al. Patient mortality is associated with staff resources and workload in the ICU: A multicenter observational study. Critical Care Medicine 2015;43(8): 1587-94. 80. Deshmukh A, Pant S, Kumar G, et al. Comparison of outcomes of weekend versus weekday admissions for atrial fibrillation. The American Journal of Cardiology 2012;110(2):208-11. 81. Bejot Y, Aboa‐Eboulé C, Jacquin A, et al. Stroke care organization overcomes the deleterious ‘weekend effect’on 1‐month stroke mortality: A population‐based study. European Journal of Neurology 2013;20(8):1177-83. 82. Glaser R, Naidu SS, Selzer F, et al. Factors associated with poorer prognosis for patients undergoing primary percutaneous coronary intervention during off-hours: biology or systems failure Journal of Cardiovascular Interventions 2008;1(6):681-8. 83. Sorita A, Lennon RJ, Haydour Q, et al. Off-hour admission and outcomes for patients with acute myocardial infarction undergoing percutaneous coronary interventions. American Heart Journal 2014;169(1):62-8. 84. Uusaro A, Kari A, Ruokonen E. The effects of ICU admission and discharge times on mortality in Finland. Intensive Care Medicine 2003;29(12):2144-8. 85. Barnett MJ, Kaboli PJ, Sirio CA, et al. Day of the week of intensive care admission and patient outcomes: A multisite regional evaluation. Medical Care 2002;40(6):530-9. 86. Schneider EB, Hirani SA, Hambridge HL, et al. Beating the weekend trend: Increased mortality in older adult Traumatic Brain Injury (TBI) patients admitted on weekends. Journal of Surgical Research 2012;177(2):295-300. 87. Turrentine FE, Wang H, Young JS, et al. What is the safety of nonemergent operative procedures performed at night? A study of 10,426 operations at an academic tertiary care hospital using the American College of Surgeons national surgical quality program improvement database. The Journal of Trauma 2010;69(2):313-9. 88. Worni M, Schudel IM, Ostbye T, et al. Worse outcomes in patients undergoing urgent surgery for left-sided diverticulitis admitted on weekends vs weekdays: a population-based study of 31 832 patients. Archives of Surgery 2012;147(7):649- 55. 89. Ahmed A, Armstrong M, Robertson I, et al. Upper GI bleeding in Scotland 2000-2010: Improving outcome but a significant weekend effect. Gut 2012;61:(S2):a61. 90. Hoh B, Chi YY, Waters M, et al. Effect of weekend compared to weekday stroke admission on thrombolytic use, in- hospital mortality, discharge disposition, hospital charges, and length of stay in the nationwide inpatient sample database: 2002-2007. Stroke 2011;42(3):158. 91. Myers RP, Kaplan GG, Shaheen AM. The effect of weekend versus weekday admission on outcomes of esophageal variceal hemorrhage. Canadian Journal of Gastroenterology 2009;23(7):495-501. 92. Orandi BJ, Selvarajah S, Orion KC, et al. Outcomes of nonelective weekend admissions for lower extremity ischemia. Journal of Vascular Surgery 2014;60(6):1572-79. 93. Tadisina KK, Chopra K, Singh DP. The “weekend effect” in plastic surgery: Analyzing weekday versus weekend admissions in body contouring procedures from 2000 to 2010. Aesthetic Surgery Journal 2015;35(8):995-98. 94. Taira BR, Meng H, Goodman MS, et al. Does “off-hours” admission affect burn patient outcome? Burns:Journal of the International Society for Burn Injuries 2009;35(8):1092-96. 95. Martin S, Street A, Han L, et al. Have hospital readmissions increased in the face of reductions in length of stay? Evidence from England. Health Policy 2016;120(1):89-99. 96. Meacock R, Doran T, Sutton M. What are the costs and benefits of providing comprehensive seven‐day services for emergency hospital admissions? Health Economics 2015;24(8):907-12. 97. Hinds AM, Ahmad D, Lopez KT, et al. Weekday v. Weekend Endoscopy:Is there a true difference in patient outcomes? A meta-analysis. WJMA. 2014:29-35.
54 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 98. Barba R, Losa J, Velasco M, et al. Mortality among adult patients admitted to the hospital on weekends. European Journal of Internal Medicine 2006;17(5):322-24. 99. Gould JB, Qin C, Marks AR, et al. Neonatal mortality in weekend vs weekday births. JAMA 2003;289(22):2958-62. 100. Laupland KB, Shahpori R, Kirkpatrick AW, et al. Hospital mortality among adults admitted to and discharged from intensive care on weekends and evenings. Journal of Critical Care 2008;23(3):317-24. 101. Laupland KB, Misset B, Souweine B, et al. Mortality associated with timing of admission to and discharge from ICU: A retrospective cohort study. BMC Health Services Research 2011;11. 102. Lee JH, Jang SY, Kang JK, et al. Improved prognosis of weekends/holidays admission for acute myocardial infarction and a decreasing weekend-effect from 2005 to 2010. European Heart Journal 2012;33:843-4. 103. Luyt C-E, Combes A, Aegerter P, et al. Mortality among patients admitted to Intensive Care Units during weekday day shifts compared with “off” hours. Critical Care Medicine 2007;35(1):3-11. 104. Meyer T, Auracher M, Heeg K, et al. Does cumulating endurance training at the weekends impair training effectiveness? European Journal of Cardiovascular Prevention and Rehabilitation 2006;13(4):578-84. 105. Meynaar IA, van der Spoel JL, Rommes JH, et al. Off hour admission to an intensivist-led ICU is not associated with increased mortality. Critical Care 2009;13(3):r84. doi:10.1186/cc7904 106. Saposnik G, Hill MD, O’Donnell M, et al. Variables associated with 7-day, 30-day, and 1-year fatality after ischemic stroke. Stroke 2008;39(8):2318-24. 107. Schilling PL, Campbell Jr DA, Englesbe MJ, et al. A comparison of in-hospital mortality risk conferred by high hospital occupancy, differences in nurse staffing levels, weekend admission, and seasonal influenza. MedicalCare 2010;48(3):224-32. 108. Wunsch H, Mapstone J, Brady T, et al. Hospital mortality associated with day and time of admission to Intensive Care Units. Intensive Care Medicine 2004;30(5):895-901. 109. Youn YH, Park YJ, Kim JH, et al. Weekend and nighttime effect on the prognosis of peptic ulcer bleeding. World Journal Gastroenterology 2012;18(27):3578-84. 110. Berger A, Stauffer J-C, Radovanovic D, et al. Comparison of in-hospital mortality for acute myocardial infarction in Switzerland with admission during routine duty hours versus admission during out of hours (insight into the AMIS plus registry). The American Journal of Cardiology 2008;101(4):422-27. 111. Bhonagiri D, Pilcher DV, Bailey MJ. Increased mortality associated with after-hours and weekend admission to the intensive care unit: A retrospective analysis. Medical Journal of Australia 2011;194(6):287-92. 112. Hansen KW, Hvelplund A, Abildstrøm SZ, et al. Prognosis and treatment in patients admitted with acute myocardial infarction on weekends and weekdays from 1997 to 2009. International Journal of Cardiology 2013;168(2):1167-73. 113. Hong JS, Kang HC, Lee SH. Comparison of case fatality rates for acute myocardial infarction in weekday vs weekend admissions in South Korea. Circulation Journal 2010;74(3):496-502. 114. Jiang F, Zhang JH, Qin X. “Weekend Effects” in patients with intracerebral hemorrhage. Intracerebral Hemorrhage Research 2011;111:333-6. 115. Keatinge WR, Donaldson GC. Changes in mortalities and hospital admissions associated with holidays and respiratory illness: Implications for medical services. Journal of Evaluation in Clinical Practice 2005;11(3):275-81. 116. Khera S, Kolte D, Palaniswamy C, et al. ST-elevation myocardial infarction in the elderly—temporal trends in incidence, utilization of percutaneous coronary intervention and outcomes in the United States. International Journal of Cardiology 2013;168(4):3683-90. 117. Mikulich O, Callaly E, Bennett K, et al. The increased mortality associated with a weekend emergency admission is due to increased illness severity and altered case-mix. Acute Medicine 2010;10(4):182-87. 118. Mitra B, Cameron PA, Fitzgerald MC, et al. “After-hours” staffing of trauma centres and outcomes among patients presenting with acute traumatic coagulopathy. Medical Journal of Australia 2014;201(10):588-91. 119. Niewada M, Jezierska-Ostapczuk A, Skowronska M, et al. Weekend versus weekday admissions in Polish stroke centres - could admission day affect prognosis in Polish ischaemic stroke patients? Neurologia i Neurochirurgia Polska 2012;46(1):15-21. 120. Numa A, Williams G, Awad J, et al. After-hours admissions are not associated with increased risk-adjusted mortality in pediatric intensive care. Intensive care medicine. 2008;34(1):148-51. 121. O’Brien EC, Rose KM, Shahar E, et al. Stroke mortality, clinical presentation and day of arrival: The Atherosclerosis Risk in Communities (ARIC) study. Stroke Research and Treatment 2011 doi: 10.4061/2011/383012 122. Ogbu UC, Westert GP, Slobbe LC, et al. A multifaceted look at time of admission and its impact on case-fatality among a cohort of ischaemic stroke patients. Journal of Neurology, Neurosurgery & Psychiatry 2011;82(1):8-13.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 55 123. Ono Y, Ishida T, Iwasaki Y, et al. The off-hour effect on trauma patients requiring subspecialty intervention at a community hospital in Japan: A retrospective cohort study. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2015;23(1):1. 124. Powell ES, Khare RK, Courtney DM, et al. The weekend effect for patients with sepsis presenting to the emergency department. The Journal of Emergency Medicine 2013;45(5):641-48. 125. Saposnik G, Baibergenova A, Bayer N, et al. Weekends: A dangerous time for having a stroke? Stroke 2007;38(4):1211- 15. 126. Smith EE, Shobha N, Dai D, et al. Risk score for in-hospital ischemic stroke mortality derived and validated within the Get With The Guidelines-Stroke Program. Circulation 2010;122(15):1496-504. 127. Smith S, Allan A, Greenlaw N, et al. Emergency medical admissions, deaths at weekends and the public holiday effect. Cohort study. Emergency Medicine Journal 2014;31:30-4. 128. Al-Lawati JA, Al-Zakwani I, Sulaiman K, et al. Weekend versus weekday, morning versus evening admission in relationship to mortality in acute coronary syndrome patients in 6 middle eastern countries: Results from Gulf Race 2 Registry. Cardiovascular Medicine Journal 2012;6:106. 129. Almekhlafi MA, Hockley A, Desai JA, et al. Overcoming the evening/weekend effects on time delays and outcomes of endovascular stroke therapy: the Calgary Stroke Program experience. Journal of Neurointerventional Surgery 2014;6(10):729-32. 130. Arias Y, Taylor DS, Marcin JP. Association between evening admissions and higher mortality rates in the Pediatric Intensive Care Unit. Pediatrics 2004;113(6):530-4. 131. Brims FJ, Asiimwe A, Andrews NP, et al. Weekend admission and mortality from acute exacerbations of chronic obstructive pulmonary disease in winter. Clinical Medicine 2011;11(4):334-9. 132. Busse J, Bhandari M, Devereaux P. The impact of time of admission on major complications and mortality in patients undergoing emergency trauma surgery. Acta Orthopaedica 2004;75(3):333-8. 133. Carmody IC, Romero J, Velmahos GC. Day for night: Should we staff a trauma center like a nightclub? The American Surgeon 2002;68(12):1048-51. 134. Desai V, Gonda D, Ryan SL, et al. The effect of weekend and after-hours surgery on morbidity and mortality rates in pediatric neurosurgery patients. Journal of Neurosurgery Pediatrics 2015;16(6):726-31. 135. Karliński M, Kobayashi A, Sobolewski P, et al. Is there a bad time for intravenous thrombolysis? the experience of Polish Stroke Centers. Neurologia i Neurochirurgia Polska 2014;48(1):45-51. 136. Koike S, Tanabe S, Ogawa T, et al. Effect of time and day of admission on 1-month survival and neurologically favourable 1-month survival in out-of-hospital cardiopulmonary arrest patients. Resuscitation 2011;82(7):863-68. 137. Kuijsten HAJM, Brinkman S, Meynaar IA, et al. Hospital mortality is associated with ICU admission time. Intensive Care Medicine 2010;36(10):1765-71. 138. Laupland KB, Kibsey PC, Galbraith JC. Community-onset bloodstream infection during the ‘after hours’ is not associated with an increased risk for death. Canadian Journal of Infectious Diseases and Medical Microbiology 2012;23(4):170-72. 139. Vest-Hansen B, Riis AH, Sørensen HT, et al. Out-of-hours and weekend admissions to Danish Medical Departments: Admission rates and 30-day mortality for 20 common medical conditions. BMJ Open 2015;5(3):e006731. doi:10.1136/ bmjopen-2014-006781 140. Goldstein SD, Papandria DJ, Aboagye J, et al. The “weekend effect” in pediatric surgery—increased mortality for children undergoing urgent surgery during the weekend. Journal of Pediatric Surgery 2014;49(7):1087-91. 141. Marco J, Barba R, Plaza S, et al. Analysis of the mortality of patients admitted to internal medicine wards over the weekend. American Journal of Medical Quality 2010;25(4):312-18. 142. Schmid M, Ghani KR, Choueiri TK, et al. An evaluation of the ‘weekend effect’ in patients admitted with metastatic prostate cancer. BJU International 2015;116(6):911-19. 143. Tung Y-C, Chang G-M, Chen Y-H. Associations of physician volume and weekend admissions with ischemic stroke outcome in Taiwan: A nationwide population-based study. Medical Care 2009;47(9):1018-25. 144. Turin TC, Kita Y, Rumana N, et al. Case fatality of stroke and day of the week: Is the weekend effect an artifact? Cerebrovascular Diseases 2008;26(6):606-11. 145. Fonarow GC, Abraham WT, Albert NM, et al. Day of admission and clinical outcomes for patients hospitalized for heart failure: Findings from the Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure (OPTIMIZE-HF). Circulation Heart Failure 2008;1(1):50-7. 146. Hamilton P, Restrepo E. Weekend birth and higher neonatal mortality: A problem of patient acuity or quality of care? Journal of Obstetric, Gynecologic & Neonatal Nursing 2003;32(6):724-33.
56 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 147. Maggs F, Mallet M. Mortality in out-of-hours emergency medical admissions - more than just a weekend effect. Journal of the Royal College of Physicians of Edinburgh 2010;40(2):115-8. 148. Arbabi S, Jurkovich GJ, Wahl WL, et al. Effect of patient load on trauma outcomes in a level 1 Trauma Center. Journal of Trauma - Injury, Infection and Critical Care 2005;59(4):815-20. 149. Auger KA, Davis MM. Pediatric weekend admission and increased unplanned readmission rates. Journal of Hospital Medicine 2015;10(11):743-5. 150. Boylan M, Rosenbaum J, Adler A, et al. Hip Fracture and the weekend effect: Does weekend admission affect patient outcomes? American Journal of Orthopedics 2015;44(10):458-64. 151. Busl KM, Prabhakaran S. Predictors of mortality in nontraumatic subdural hematoma: Clinical article. Journal of neurosurgery 2013;119(5):1296-301. 152. Chang G-M, Tung Y-C. Factors associated with pneumonia outcomes: a nationwide population-based study over the 1997–2008 period. Journal of General Internal Medicine 2012;27(5): 527-33. 153. Suissa S, Dell’Aniello S, Suissa D, et al. Friday and weekend hospital stays: Effects on mortality. European Respiratory Journal 2014;44(3):627-33. 154. Thomas C, Smith R, Uzoigwe C, et al. The weekend effect. Bone Joint Journal 2014;96(3):373-8. 155. Zapf MA, Kothari AN, Markossian T, et al. The “weekend effect” in urgent general operative procedures. Surgery 2015;158(2):508-14. 156. Zhang G, Zhang JH, Qin X. Effect of weekend admission on in-hospital mortality after subarachnoid hemorrhage in Chongqing China. 2011:229-32. 157. Kwan JL, Bell CM. Should we rethink the scheduling of elective surgery at the weekend? Higher 30-day mortality for elective procedures scheduled Friday through Sunday. BMJ (Online) 2013;346:f3353. doi:10.1136/bmj.f3353 158. Fang J, Kapral MK, Saposnik G, et al. Higher mortality rates in patients presenting to hospital with acute stroke on weekends maybe explained by higher initial stroke severity. Stroke 2009;40(4):183. 159. Sharp AL, Choi H, Hayward RA. Don’t get sick on the weekend: An evaluation of the weekend effect on mortality for patients visiting US EDs. The American Journal of Emergency Medicine 2013;31(5):835-7.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 57 Appendices
Appendix 1: Project Plan Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project
Project Title
A review of health outcomes for patients admitted after hours and at the weekend: Is there an effect and what are the characteristics and contributing factors?
Reviewers
Associate Professor Anthony Bell: SOM UQ & Medical Director, Department of Emergency Medicine, Royal Brisbane and Women’s Hospital.
Dr Roger O’Gorman: Medical Officer, Department of Emergency Medicine, Royal Brisbane and Women’s Hospital. Ms Michele Romeo: Principal Project Officer, Office of Chief Executive Metro North Hospital and Health Services.
Beginning and Ending Dates
10 December 2015–30 June 2016.
Definition
Out of hours: out of regular working hours (off-hours), during weekends, during holiday periods, on public holidays and on evening and night-time shifts on weekdays.
Introduction
Over the last decade, an increasing number of research studies have examined the association between weekend hospital admissions and poorer patient outcomes including higher rates of mortality. There is significant evidence demonstrating inequality in health outcomes for patients admitted out of hours.
A range of potential causal links for inequitable outcomes for patients admitted out of hours have been identified, one of these is the availability of staff and services at weekend. It remains unclear how many deaths are avoidable. While further evidence on the causal relationships would be of value, we do not think that this can be a justification for taking no action where there are good clinical reasons for doing so.
Background information
Historically, many healthcare facilities provide all services Monday to Friday, with these reducing skeleton staff out of hours and on the weekend. Whilst the nature of human illness is not determined by time of day or day of week, Queensland Public Health Services currently structure health service delivery in this way around a five day delivery model. It has been documented for in the literature that admission to hospital out of hours and on the weekend is associated with poor patient outcomes. Potentially, this presents inconsistency between demand (patients requiring healthcare seven days a week) and supply (healthcare facilities delivery a full complement of health services five days a week). Internationally a number of countries are now endeavouring to develop a systems based approach to planning a transition from five to seven day health care delivery models and some services independent instituting program reorganisation to achieve these ends as research (and expert opinion) highlights increase mortality and morbidity for weekend and after hours admission to hospitals.
There is a growing body of evidence that morbidity and mortality rates increase for patients admitted to hospital on the out of hours. While researchers continue to argue about cost effective ways of improving health outcomes on the weekend, moves are being made globally and locally to provide consistent access to comprehensive healthcare across the seven day week. Key research papers and reviews in this area to date are:
Freemantle et al. (2015). Increased mortality associated with weekend hospital admission: case for expanded 7-day services. BMJ. (http://www.bmj.com/content/351/bmj.h4596)
58 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project From analyses 2013-14 Hospital Episode Statistics data it was found that:
• Although there are fewer hospital admissions at weekends, patients who are admitted on Saturday and Sunday are sicker and face an increased likelihood of death within 30-days, even when severity of illness is taken into account.
• Patients admitted on a Saturday or Sunday have a 15% and 10% respectively greater risk of mortality compared to those admitted on Wednesday.
• There are approximately 11,000 excess deaths in-hospitals every year among patients admitted on a Friday, Saturday, Sunday or Monday compared with other days of the week. The authors included the effect of Fridays and Monday as “appropriate support services in-hospitals are usually reduced from later Friday through the weekend, leading to disruption on Monday morning.”
• Oncology patients admitted on a Sunday have a 29% increased risk of death compared to those admitted on a Wednesday.
• Patients with cardiovascular disease admitted on a Sunday have a 20% increased risk of death compared to those admitted on a Wednesday.
• The research concluded that it is not possible to determine how many of the excess deaths were avoidable, but that the statistic is ‘not otherwise ignorable’ and ‘raises challenging questions about reduced service provision at weekends.’
Ruiz et al. (2015). The Global Comparators Project: International comparison of 30-day in-hospital mortality by day of the week. BMJ Quality and Safety. (http://qualitysafety.bmj.com/content/24/8/492)
• The Global Comparators dataset collected records of inpatients from 50 hospitals across 10 countries (incl: subset of 28 hospitals across England, Australia, USA and Netherlands) presenting to an emergency department showed:
• There is an overall 30-day crude mortality rate of 3.9%; the English hospitals had the highest crude morality rate (4.6%). Crude mortality rates for the English, Dutch and USA hospitals were higher at weekends compared with weekdays.
• Emergency patients in the English, USA and Dutch hospitals showed a significantly higher adjusted risk of death within 30- days following admission on a Saturday or Sunday compared with admission on a Monday.
• This study did not indicate a difference in mortality within 30-days for patients admitted at weekends in Australian hospitals. However analysis of mortality within 7-days, the Australian hospitals showed 12% higher risk of death when admitted on a Saturday compared to a Monday, and 11% higher risk of death following a Sunday admission.
East Midlands Clinical Senate (2014), 7-day Services Project: Acute Collaborative Report. (http://www.hsj.co.uk/Journals/2015/05/05/c/k/v/East-Midlands-Clinical-Senate-report.pdf)
A review of the current provision against the 10 clinical standards for urgent and emergency care that underpin consistently high quality care 7-days a week by ten East Midlands acute trusts found that:
• There is more to do for all trusts to meet the 10 clinical standards, or all of the 4 priority standards, with current performance rated as amber overall.
• There is a potential need for network arrangements between trusts to ensure 7-day access to quality care, such as by developing a larger pool of clinicians in specific specialities or services.
• While services will need to be redesigned to ensure the availability of workforce, a more fundamental culture shift is also needed to embrace the necessity to provide necessary services across 7-days.
NHS services. (2013). 7-days a week forum. (http://www.england.nhs.uk/wp-content/uploads/2013/12 /evidence-base.pdf)
NHS England’s NHS Services, Seven Days a Week Forum was a clinically-led process which included an extensive review of the published literature alongside analysis of HES data to explore patient outcomes at weekends compared to during the week.
The review found that:
• There is significant variation in patient outcomes for those admitted as an emergency. This variation is seen in patient experience, mortality rates, length of hospital stay and re-admission rates.
• There is a large body of evidence associating timely consultant input to patient care with improved outcomes.
• Radiology and endoscopy are examples of key interventions which are a time critical response to an urgent or emergency need. However, service provision is shown to be highly variable, particularly at weekends.
• Consultant-delivered ward rounds are a central pillar for patient care. However, reduced weekend service levels mean many hospitals do not meet national recommendations for twice daily consultant ward rounds.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 59 Freemantle et al. (2012). Weekend hospitalisation and additional risk of death: an analysis of inpatient data. J R Soc Med. (http://www.ncbi.nlm.nih.gov/pubmed/22307037) Analysis of 2009/10 data found:
• Patients admitted to hospital on a Sunday had a 16% greater risk of death within 30-days compared to those admitted on a Wednesday.
• Patients admitted on a Saturday had an 11% increased risk of death within 30-days compared to those admitted on a Wednesday.
• Day of admission was associated with increased risk of death in 7 of the 10 most common CCS groups (clinical conditions). For example:
• Patients admitted on a Sunday with acute and unspecified renal failure had a 37% increased risk of death compared with those admitted on a Wednesday
• Patients admitted on a Sunday with acute myocardial infarction had an 11% increased risk of death compared to those admitted on a Wednesday
AoMRC. (2012). 7-day consultant present care. (http://www.aomrc.org.uk/doc_view/9532-seven- day-consultant-present-care)
In light of evidence demonstrating less favourable patient outcomes at weekends compared to weekdays, the Academy of Medical Royal Colleges presented proposals for achieving parity for inpatient care throughout the week. The report indicated:
• Most hospitals and specialities already provide a non-resident consultant-led on-call rota which should ensure acutely unwell or deteriorating patients have access to consultants and timely interventions. However, in the absence of daily ‘planned’ consultant review, the remainder of the patient’s care pathway is often put into hibernation particularly over weekends, resulting in delays in diagnosis, investigation, treatment and discharge from hospital.
• It is not uncommon for patients whose condition is not deteriorating to wait until the next scheduled weekday review before being seen by a consultant. For example, a patient who is admitted on a Thursday night will usually be seen by a consultant on Friday morning, but may then wait until Monday for their next scheduled consultant review.
• The weekend effect is very likely attributable to deficiencies in care processes linked to the absence of skilled and empowered senior staff in a system which is not configured to provide full diagnostic and support services 7-days a week.
• Following discharge from acute areas to general wards the frequency of consultant review falls significantly. The result is that departures from the care pathway are not uncommon, and are not detected in a timely manner.
• The most effective way to improve outcomes for patients admitted to hospital at weekends is to ensure that care is delivered by adequately supported consultants and monitored during care pathways.
Aylin et al. (2010). Weekend mortality for emergency admissions: a large multicentre study. BMJ Quality and Safety. (http://qualitysafety.bmj.com/content/19/3/213.short)
• This study is one of the first, large scale studies of English data to explore weekend mortality rates for emergency admissions. From 2005/6 data, the study found:
• Crude mortality rates are higher for patients admitted at weekends compared to weekdays (5.2% for all weekend admissions; 4.9% for all weekday admissions; overall crude mortality rate: 5.0%). There is a 10% higher risk of death for patients admitted as an emergency at the weekend compared with those admitted on a weekday.
• There may be a possible 3,369 excess deaths occurring at the weekend compared to weekdays (equivalent to a 7% higher risk of death).
• In addition to the evidence of weekend hospital admissions resulting in poorer patient outcomes including higher rates of mortality, there is also some evidence that trainee healthcare staff are also affected by existing weekend practices. A reduction in available senior support at the weekend can lead to an ineffective use of time to support training and improve skills for the benefit of patients and staff. As discussed in the report below.
Professor Sir John Temple. (2010). Time for training. (http://hee.nhs.uk/healtheducationengland/files /2012/08/Time-for-training-report.pdf)
60 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project This report reviewed the impact of the European Working Time Directive (EWTD) on quality of training for doctors, dentists, healthcare scientists and pharmacists. The report found:
• Rigid, poorly designed rotas can result in trainees being unsupported and unsupervised
• Splitting services into elective and emergency can enhance training, deliver EWTD-compliant rotas and improve quality of care.
• Elective work is relatively time driven and proactive and provides good speciality training. By contrast, emergency work is variable over 24 hours (but with predictable peaks) and can provide valuable training but is often not maximised as many trainees are unsupported and poorly supervised.
There is little support for extending hours or lengthening training programmes. This can perpetuate the situation of staff on specialty training programmes not using the hours effectively for training, which can lead to trainees providing the frontline services outside of hours, unsupported and without direct supervision.
Whilst the evidence for health outcomes for admissions out of hours is mainly observational not all research studies have reported inequality of health at these times. Factors that have been proposed to explain the difference in health outcomes for admissions out of hours are staffing issues, effective hand-over communication and the delay in diagnostic and therapeutic procedures.
To date very little research has been undertaken examining the effectiveness of interventions to improve outcomes out of hours with expert option stressing the need for well-planned trials to obtain good quality evidence of the effectiveness of interventions to improve patient outcomes at the weekend (P. Alyin, personal communication September 11, 2015).
The QCS 2015 March meeting explored the issues and opportunities associated with out of hours and weekend service delivery system in Queensland. QCS recommended that essential services that are available on weekdays should also be available out of hours and at the weekend along with improving to access to care and communication between healthcare professionals.
The QCS endorsed for an evidenced based review to explore how diagnostic groups and causal pathways impact of admissions out of hours in Queensland to inform diagnosis and service specific solutions and investment priorities for HHS in Queensland.
Aim
Identify the underlying causes of differential health outcomes of patients admitted outside of hours, which includes weekends, and explore intervention/s that have been show to improve outcomes for patients admitted at these times.
Objectives
• This study is a descriptive literature review that addresses the issue of inequitable health outcomes that occur from admissions out of hours. The specific objectives include:
• To identify and analyse the factors that influence inequality of health outcomes to patients admitted/presenting to hospital out of hours and at the weekend.
• To identify the range and scope of services specifically targeting patients admitted /presenting to hospital out of hours and at the weekend.
• To identify models of care that has been demonstrated to reduce the inequality of health outcomes for patients admitted/ presenting to hospital out of hours and at the weekend and evaluate their effectiveness.
• To identify the policy implications and identify areas requiring further research.
Condition or domain being studied
• Participants/ Population: The population is patients admitted out of hours.
• Exposure(s): The exposure is defined as an intervention that addresses the inequality of healthcare outcomes for patients admitted out of hours.
• Comparator(s)/control: Patients admitted on-hours (within regular working hours).
• Outcome(s): The outcomes of interest are short-time mortality, patient outcomes other than mortality (such as functional recovery or readmission) and models of care/ interventions type to encourage equitable health outcomes for patients admitted after hours and at the weekend.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 61 Methods of the review
Because of the broad nature of the review topic, a comprehensive search strategy has been developed and will be cycled through a number of iterations in order to maximise comprehensiveness and precision.
Stage CI: PPO: CE: Define Study Question Approve Lead Advice Prepare Protocol Review Lead Review Undertake Search Stage 1: Title Search -Relevance Lead Data Extraction Stage 2: Abstract Review- Significance Review Lead Lead Quality Assessment Stage 3: Full Text Review -Impact Review Lead Lead Data Synthesis Review Lead Lead Interpretation Review Lead Lead Write Report Review Lead Review
Search strategy for identification of studies
Initially, scoping search was performed to identify major papers on published evidence and refine the final key terms used to undertake the search strategy detailed below.
The following concepts and all possible synonyms (search terms) are considered:
Concept: out of hours and at the weekend: 24/7 model; 7-day care; 7-day health; 7-day services; after hours; day of week; late in week; night care; night time; night time; off-hours; off shift ;out of hours; round the clock; utilisation; weekend; weekend service.
Concept: hospital & departments: acute; allied health; care; chemotherapy; elective; emergency department; endoscopy; health care; hospital; intensive care; inter cranial haemorrhage; medical; microbiology; myocardial infarction; nursing; operating theatre; pathology; PCI; radiology; rehabilitation; service/s; STI; stroke; surgery/surgical; trauma; unit; unscheduled; unscheduled; ward.
Concept: study design: Cohort study/studies; Longitudinal study/studies; Follow up study/studies; Prospective study/studies; Retrospective study/studies; Prevalence study/studies; Incidence study/studies; transversal.
Concept: outcomes: Assessment; clinical governance; co morbidity; criteria led discharge; discharge; equity; governance; intervention; medical error; models of care; morbidity; mortality; patient harm; patient safety.
Types of study to be included/ excluded
In this type of research the gold-standard of an RCT is often impossible and sometimes an inappropriate technique. As such all designs of this study with appropriate outcomes measures will be considered including: RCTs, quasi-experimental studies, cohort studies, cross-sectional studies observational studies /descriptive papers (where higher quality of evidence is not available). Also written material or information that is unpublished or not published commercially: that which is produced on all levels of government, academics, business and industry, in print and electronic formats, but which is not controlled by commercial publishers” (grey literature) including: (conference papers, blogs, fact-sheets, dissertations, newsletters, course materials, book chapters, memoranda, annual reports, interviews, policy statements, posters, government documents, newspaper clippings, personal communication, pamphlets, essays, emails, questionnaires, speeches, press releases, speeches). This information provides a broad range of information and contains new ideas.
Articles will automatically be excluded if it is published in a language other than English; and/or includes data from a hospital system in an under-developed country.
Articles will be peer reviewed from a reputable source or have been contributed by experts in the field of medicine, nursing, patient safety, health policy, health system economics and /or finance. Additionally, the articles will be those which investigators feel that make a significant, timely or unique contribution to the development of our understanding in this research area.
62 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project Data Extraction (selection and coding)
The data extraction process will be performed by two reviewers independently. In case of disagreement, consensus will be sought by discussion and in case of disagreement a third party will be consulted.
A number of approaches will be utilised and involve 3 stages as follows:
• Title (eye-balling) Search. Titles will be screened for relevance (including check for duplicates) to the aims of the study.
• Abstract Review - Abstracts will be screened for significance (& utility) to the aims of the study.
• Full Text Review - Full text format will be analysed for their contribution (impact) that may meet the objectives of the study. Stage 1: Title search
Screen title for relevance to the project. Articles to be included if they appear to relate to the management of patients admitted/ presenting to hospital out of hours and in particular to models of care and the factors that appear to be driving changes in those models. Studies that are clearly not relevant to the aims of the study will be excluded without any further attention.
Reference titles are to be scanned to identify contributions to the definition, criteria, components and conceptual framework of models of care targeting the identified causes of inequitable outcomes for patients admitted out of hours.
At this stage no limitations are to be placed on publication date in order to retrieve the history and scope of the project topic.
Search of key bibliographic databases using a set of key search combinations including:
• CINAHL (incl. PsycINFO).
• Cochrane Library.
• EMBASE.
• Google scholar.
• Joanna Briggs Institute Evidence Based Practice database.
• PubMed (and MedLine).
• Scopus.
Focused review of the contents of national and international medicine, quality safety, health and economic journals, grey literature (including guidelines and unpublished research) and thesis with the assistance of:
• ‘Advanced’ Google search.
• ProQuest Dissertations and Theses Global.
• Open Grey.
• Grey Matters.
Additional references will be identified through examination of references manually from the most recent publications from specific journals, trial registers, specific websites (including professional organisation pages) (snowballing) and through scrutiny of contents pages of highly relevant journals for the last three years including:
British Medical Journal (BMJ); Journal of America Medical Association (JAMA); New England Journal of Medicine; BMJ Quality & Safety; Critical Care Medical Journal; Medical Journal of Australia (MJA); The Lancet and the Canadian Medical Association Journal. Stage 2: Abstract review
Second screen for relevance to the project: screening for impact.
Articles to included are those that appear to make a significant contribution of information that is the article provide evidence to support the background (content and scope) as well as the factors contributing to inequality of health for people admitted out of hours, the analysis of the strategies/ interventions and models of care to affect the inequality of health for patients admitted/presenting to the hospital out of hours.
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 63 Studies that are excluded at this stage will be recorded with a short explanation for the exclusion. Reviewers will assess the abstracts based on the following:
CATEGORY 1 Publications that contain context and background.
Publications that detail existing interventions/models of care, their content CATEGORY 2 and scope.
CATEGORY 3 Publications that provide recommendations for interventions/models of care that are not evidence based.
CATEGORY 4 Publications that provide recommendations for interventions/models of care that are evidence based.
Publications that provide recommendations for interventions/ models of care that are evidence-based and CATEGORY 5 derived from appropriately structured research methods
CATEGORY 6 Other relevant publications that do not align with categories 1-4
Stage 3: Full text review
Final screen for significance to the project.
Articles to be included are those that appear to be strong in the articles underlying logic and the investigators are confident that conclusions can be drawn. Articles to be included are those that make a significant contribution.
The assessment of the evidence will be performed by two reviewers independently. In case of disagreement consensus will be sought by discussion, and in case of further disagreement a third party will be consulted
Reviewers will assess the full text format for their contribution towards objectives of the project. Reviewers will assess the full text based on the following:
Level 1 Systematic Review/ Meta-analysis RCTs Experimental Level 2 Cohort control studies Case control studies ‘Outcomes’ research Observational research Level 3 Case studies Other types of studies (e.g., interview based, local audit) Level 4 Expert opinion without explicit critical appraisal
Risk of bias (quality) assessment: The included studies will be thoroughly read and data will be extracted following a pre- defined data extraction template which will consider the appropriateness of study design to the research objective, risk of bias, and other issues including (choice of outcome measure, statistical issues, quality of reporting, quality of the intervention, generalisability).
Strategy for data synthesis – Interpretation (final report)
A textual narrative synthesis of these studies exploring the relationships between the studies and (if appropriate) will combine results from studies included in the review. The review will include an analysis of the individual selected studies their characteristics, findings and level of quality (including heterogeneity) will be collated, combined and summarised be presented in a diagram (table) reviewing and reporting key information including study setting, year of study, study design, quality of the design (strength/weakness), major findings and implications.
Contact details for further information
Michele Romeo, Principal Project Officer - REACH.
Office of the Chief Executive | Metro North Hospital and Health Service | Royal Brisbane and omen’W s Hospital Ground Floor, Dr James Mayne Building, Butterfield Street, Herston QLD 4006. [email protected]
64 Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project Appendix 2: Search Strategies
Database Date Range Search Date Search Strategy Results
Scopus 1995-01-01 to 2015-12-22 (hospital* OR “emergency health service*” OR “emergency health 3,876 2015-12-22 service” OR “emergency department*” OR “emergency department”) AND (“7-day” OR “seven day” OR “night care” OR “after-hours care” OR “week end” OR weekend* OR “after hours” OR “out of hours”) PUBMED 1995-01-01 to 2015-12-11 (hospital:de,ab,ti OR hospitals:ab,ti OR ‘emergency health service’/ 2,347 2015-12-11 exp OR ‘emergency health service’ OR ‘emergency departments’:ab,ti OR ‘emergency department’:ab,ti) AND (‘7-day’:ab,ti OR ‘seven day’:ab,ti OR ‘night care’:ab,ti OR ‘after-hours care’:ab,ti OR ‘week end’:ab,ti OR weekend:ab,ti OR ‘after hours’:ab,ti OR ‘out of hours’:ab,ti) AND (‘cohort analysis’/exp OR ‘longitudinal study’/ exp OR ‘follow up’/exp OR ‘prospective study’/exp OR ‘retrospective study’/exp OR cohort:ab,ti OR longitudinal:ab,ti OR prospective:ab,ti OR retrospective:ab,ti OR audit:ab,ti OR epidemiology:ab,ti OR ‘cross-sectional study’/exp OR ‘cross-sectional’:ab,ti OR ‘prevalence study’:ab,ti OR ‘prevalence studies’:ab,ti OR ‘incidence study’:ab,ti OR ‘incidence studies’:ab,ti OR ‘transversal study’:ab,ti OR ‘transversal studies’:ab,ti) EMBASE 1995-01-01 to 2015-12-22 (hospital OR hospitals OR “emergency health service” OR 4,122 2015-12-22 “emergency health services” OR “emergency departments” OR “emergency department”) AND (“7-day” OR “seven day” OR “night care” OR “after-hours care” OR “week end” OR weekend* OR “after hours” OR “out of hours”) Emerald - Insight 1995-01-01 to 2015-12-21 (hospital* OR “emergency health service*” OR “emergency health 341 2015-12-21 service” OR “emergency department*” OR “emergency department”) AND (“7-day” OR “seven day” OR “night care” OR “after-hours care” OR “week end” OR weekend* OR “after hours” OR “out of hours”) CINAHL 1995-01-01 to 2015-12-21 (hospital* OR “emergency health service*” OR “emergency health 1,816 2015-12-21 service” OR “emergency department*” OR “emergency department”) AND (“7-day” OR “seven day” OR “night care” OR “after-hours care” OR “week end” OR weekend* OR “after hours” OR “out of hours”)
Reviewing Equitable Access to healthCare outcomes out of Hours and at the weekend (REACH) Project 65 66 Appendix 3: Data extraction summary of studies reporting the outcomes for patients admitted out of hours and at the weekend - Mortality Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project
Disease specific In- / non-disease 2-days 3-days 5-days 7-days 14-days 15-days 28-days 30-days 60-days 90-days 120-days 365-days hospital specific type Medical 10 4 6 12 22 4 24 98 124 141 6 98 124 147 1 16 10 24 11 117 37 107 127 139
Surgical 6 14 15 32 87 4 7 17 31 32 1 31 1 31 93 155 Intensive Care 15 21 67 111 26 1 130 1 111 28 29 33 78 79 84
100 101 103 105
108 120 137
Cardio Respiratory 18 34 44 52 59 1 115 9 50 62 65 66 73 1 145 1 145 1 128 63 65 66 69 80 82 113 128 136 152
83 92 110 112 115 116 131 145 153 Nephrology 1 13 1 13
Neuroscience 23 19 20 75 76 1 68 9 46 53 68 75 76 3 75 76 125 1 121 9 55 68 75 76 81 2 71 129 151 46 48 58 90 122 125 135 144 125 134 143 144 114 36 60 64 72 86 119 125 126
143 156 159
Gastroenterology 12 35 38 39 4 27 41 74 89 41-43 45 49 54 74
88 109
Haematology - 4 40 47 138 142 1 40 2 40 47 Oncology and other medical specialties Paediatrics and 5 18 99 140 Obstetrics 146 149 Trauma 10 21 56 70 94 132 133 123 Orthopaedics 4 57 61 150 154 1 77 4 51 61 77 154 1 61 Appendix 4: Data extraction summary of studies reporting the outcomes for patients admitted out of hours and at the weekend - Treatment Delay
Disease specific / non disease specific type Treatment delay out of hours No treatment delay out of hours Cardio Respiratory 3 44 50 52 5 59 62 63 65 66 Neuroscience 3 46 48 53 Gastroenterology 9 35 38 39 41-43 45 49 54 6 19 20 55 58 60 64 Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project Haematology-Oncology and other medical specialties 2 40 47 Orthopaedics 1 51 2 57 61
Appendix 5: Data extraction summary of studies reporting the outcomes for patients admitted out of hours and at the weekend - Length of Stay
Disease specific / non disease specific type Difference in Length of Stay No difference in Length of Stay Medical 1 37 Surgical 1 87 1 93 Intensive Care 1 85 2 33 78 120 Cardio Respiratory 2 50 73 1 92 Neuroscience 1 86 3 19 20 90 Gastroenterology 2 54 88 5 27 35 38 89 91 Haematology-Oncology and other medical specialties 2 40 47 Trauma 3 56 94 148 Orthopaedics 1 150 67 68 Appendix 6: Data extraction summary of studies reporting the outcomes for patients admitted out of hours and at the weekends - Timeframe Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project
Weekends: all Disease specific Analysed weekends and Analysed weekends other times Dirunal Each day studied Other Total / non disease nights seperately and nights together considered specific type weekdays Yes Yes No Mixed Yes No Mixed No Yes No Mixed Yes No Mixed Yes No Mixed
ACACACACACAC CAC CAAC ACA AACA Systematic review Case series Meta analysis Systematic review Cardio Respiratory 1 69 2 52 5 59 63 2 34 3 44 50 2 66 92 3 62 1 113 2 1 2 152 2 73 80 26 52 65 128 131 116 83 112 145 153
136 110 115
Gastroenterology 1 42 2 43 74 1 49 1 1 45 1 38 1 41 2 139 1 89 3 27 1 15 109 88 91 39 Neuroscience 2 48 2 36 129 1 7 46 53 76 2 68 3 19 20 75 1 5 58 60 2 55 1 151 1 72 1 1 143 1 30 134 135 93 122 125 119 114 71 81 90 121 156 144
126 Oncology 1 138 1 47 1 142 1 40 4 Orthopaedics 1 57 1 51 1 77 1 61 1 154 1 150 6 Paediatrics 1 18 1 146 1 99 1 149 1 140 5 Nephrology and Renal Trauma 4 56 94 1 86 1 118 1 64 1 70 1 1 10 132 133 123 148
Medical 1 84 1 98 5 10-12 16 2 117 124 1 37 3 4 22 1 147 1 8 1 24 3 6 9 1 20 107 127 141 24
Surgical 1 31 3 7 15 17 1 32 1 87 1 155 2 14 93 9 Intensive Care 1 9 21 28 4 33 1 111 1 1 26 1 23 1 79 1 85 20 137 29 35 67 84 120
78 101 100
103 105 130
Total 2 7 2 21 2 6 23 4 9 10 10 6 1 1 1 1 1 1 1 7 2 1 1 3 1 1 11 2 4 3 145
*A= administrative data-set *C= clinical data-set Appendix 7: Data extraction summary of studies reporting the outcomes for patients admitted out of hours and at the weekend - Region and dataset (administrative & clinical)
Country Study Type Disease specific / non-disease specific type Care
Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project Total Renal Trauma Medical Surgical Oncology Paediatrics Orthopaedics Neuroscience Haematology- Intensive Nephrology and Gastroenterology Cardio Respiratory
Asia Administrative 2 113 152 3 117 143 144 5 Clinical 2 62 136 2 41 109 2 29 33 3 48 114 156 1 123 10 Australia Administrative 1 111 2 8 24 3 Clinical 1 120 1 36 1 118 3 Canada Administrative 2 44 153 1 54 1 138 1 4 2 46 126 7 Clinical 1 91 2 101 100 1 129 1 70 5
Middle-East Administrative Clinical 1 128 1 21 2 Middle-East Administrative Clinical 1 45 1 23 1 32 3 Other Europe Administrative 1 69 2 141 139 141 3 122 81 135 2 51 57 8
Clinical 3 52 110 112 1 42 6 67 79 84 103 1 98 1 119 1 77 13 105 137
UK Administrative 1 66 3 49 89 74 6 6 10–12 117 147 1 53 1 18 3 7 14 17 15
Clinical 4 63 65 115 131 1 108 1 37 2 55 68 2 61 154 10 USA Administrative 5 50 73 80 92 116 4 38 88 43 39 1 142 1 85 7 9 16 22 25 107 127 10 19–20 60 71 75 1 150 4 99 140 146 149 1 13 3 15 93 155 2 64 86 39 76 90 93 125 151
Clinical 5 34 59 82 83 2 35 27 2 47 40 4 78 28 26 130 3 72 121 134 2 87 31 5 56 94 132 133 148 23
Total 26 15 4 20 20 30 6 5 1 9 10 146 69 70 Appendix 8: Data extraction summary of studies reporting the outcomes for patients admitted out of hours and at the weekend – DRG and data set (administrative and clinical) Reviewing Equitable Access tohealthCareoutcomesout ofHoursandattheweekend (REACH) Project
Yes No Mixed TOTAL (Yes) TOTAL (No) TOTAL (Mixed) TOTAL Grand Total
Disease specific / non-disease specific type Care Care Care
Cardio Respiratory Gastroenterology Neuroscience Orthopaedics Trauma Medical Surgical Intensive Cardio Respiratory Gastroenterology Neuroscience Orthopaedics Paediatrics Nephrology Trauma Medical Surgical Intensive Cardio Respiratory Gastroenterology Neuroscience Haematology-Oncology Orthopaedics Paediatrics Trauma Medical Surgical Intensive
Administrative 8 44 50 4 43 9 46 53 4 18 1 13 1 10 6 6 7 2 85 45 2 66 92 3 38 3 19 2 1 57 1 64 2 14 1 1 39 7 58 2 51 5 127 1 32 17 76 69 80 54 74 76 93 146 140 86 8-12 16 14 15 111 49 89 20 75 138 117 113 60 71 150 147 24 116 152 139 122 125 149 84 107 17 93 142 124 81 90 4 22 153 126 143 141 155 135
151 144
Clinical 2 52 82 1 42 5 48 68 2 77 1 1 98 2 79 14 9 62 63 5 27 4 36 2 40 1 61 6 56 1 37 2 31 10 21 40 4 34 2 55 1 5 26 12 66 72 119 154 118 137 65 83 110 41 88 114 129 47 70 94 87 28 29 35 112 121 123 33 84
134 128 136 91 156 132 67 78 115 100
145 108 133 101 103 131 130
148 105 120
Case Series 1 99 1 1
Meta Analysis 1 45 1 1 Systematic 1 25 1 23 2 2 Review Grand Total 10 6 14 2 4 1 2 11 6 4 60 11 8 7 4 2 1 7 3 2 10 55 5 1 9 2 1 6 1 6 31 146